A substation lightning arrester monitoring fault diagnosis method based on intelligent sensor

Abstract

The present application relates to the technical field of lightning arrester monitoring and diagnosis, and relates to a substation lightning arrester monitoring and fault diagnosis method based on intelligent sensors. The present application judges whether the lightning arrester has appearance damage through the appearance image of the lightning arrester after being struck by lightning. If the lightning arrester does not have appearance damage, the current characteristic data and the current environmental data of the lightning arrester are synchronously collected, compared and analyzed with the corrected reference current characteristic data, and whether the lightning arrester is in a fault state is judged. When the lightning arrester is in a fault state, the current characteristic data and the current environmental data collected are comprehensively compared and analyzed to diagnose the specific fault type of the lightning arrester, provide a clear maintenance direction for maintenance personnel, and improve the maintenance efficiency and pertinence. Otherwise, the longitudinal comparison is performed based on the current characteristic data, and corresponding processing measures are executed based on the longitudinal comparison result, so that the equipment performance degradation trend is captured in advance, and early fault warning is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of lightning arrester monitoring and diagnosis, and relates to a substation lightning arrester monitoring fault diagnosis method based on an intelligent sensor. BACKGROUND

[0002] As a core device for resisting lightning and overvoltage in the power system, the operating state of the lightning arrester is directly related to the stability and safety of the power system. With the development of the power system towards intelligentization and remoteization, higher requirements are put forward for the real-time, accuracy and comprehensiveness of lightning arrester fault diagnosis. Lightning is one of the main causes of lightning arrester failure. After lightning, the lightning arrester may have problems such as external structural damage or internal electrical performance degradation. If it is not diagnosed in time, it may cause serious consequences such as damage to power equipment and power interruption.

[0003] In the prior art, a Chinese invention patent with publication number CN118884303B discloses a zero-impedance lightning arrester fault remote diagnosis method and system. The technology performs first round stability diagnosis by collecting initial operating data of the zero-impedance lightning arrester, evaluates lightning withstand capacity, calculates environmental and electrical interference coefficients and constructs an interference coupling model, and finally completes secondary fault diagnosis based on the time sequence set of the interference coupling factor, effectively solving the problem of high false alarm rate caused by environmental and electrical interference in the remote monitoring of the zero-impedance lightning arrester.

[0004] However, the prior art has the following problems: 1. The prior art only analyzes the electrical operating data and interference factors of the zero-impedance lightning arrester, ignoring the possible external structural damage such as shell damage and insulation layer cracking of the lightning arrester after lightning. If such damage is not found in time, it may cause distortion of subsequent electrical performance monitoring data, even cause structural failure expansion, and affect the safe operation of the equipment.

[0005] 2. The prior art has insufficient diagnosis accuracy for specific fault types. The fault level is determined by the threshold value of the interference coupling factor, but it does not combine the environmental rationality verification of the current derivative parameter and historical fault cases, cannot accurately locate the fault source, and makes it difficult for maintenance personnel to quickly develop targeted maintenance solutions, affecting maintenance efficiency.

[0006] 3. The prior art mainly relies on the data of lightning arrester in the same region to evaluate the lightning withstand capacity, lacks a longitudinal comparison mechanism for the historical data of the lightning arrester itself, does not perform trend analysis on the operating data of the lightning arrester itself at different historical periods, and is difficult to capture gradual abnormal changes in current characteristic data, and cannot realize early warning of potential chronic faults. SUMMARY

[0007] The application aims to solve the problems in the prior art and provide a substation lightning arrester monitoring fault diagnosis method based on intelligent sensors, which realizes comprehensive monitoring of the appearance and electrical performance of lightning arresters after lightning strikes, precise comparative analysis of environmental adaptation, and precise identification of specific fault types.

[0008] The application solves the technical problems by adopting the technical scheme of a substation lightning arrester monitoring fault diagnosis method based on intelligent sensors, which comprises the following steps: S1, collecting the appearance image of the substation lightning arrester after the lightning strike by the intelligent sensor, and analyzing and judging whether the lightning arrester has appearance damage based on the characteristic parameters in the appearance image.

[0009] S2, if the lightning arrester does not have appearance damage, the current characteristic data and current environmental data of the lightning arrester are synchronously collected, wherein the current characteristic data includes the total leakage current and the resistive current.

[0010] S3, the reference current characteristic data of the lightning arrester of the same type in the historical normal operation condition of the substation is called, and the reference current characteristic data is corrected based on the current environmental data.

[0011] S4, the collected current characteristic data is compared and analyzed with the corrected reference current characteristic data to determine whether the lightning arrester is in a fault state.

[0012] S5, when the lightning arrester is in a fault state, the collected current characteristic data and current environmental data are comprehensively compared and analyzed to diagnose the specific fault type of the lightning arrester.

[0013] S6, otherwise, the current characteristic data is compared with the corresponding historical current characteristic data of the lightning arrester, and corresponding processing measures are performed based on the longitudinal comparison result.

[0014] Compared with the prior art, the application has the following beneficial effects: (1) the application collects the appearance image of the substation lightning arrester after the lightning strike, analyzes and judges whether the lightning arrester has appearance damage based on the characteristic parameters in the appearance image, realizes the cooperative monitoring of the appearance state and electrical performance, avoids the subsequent diagnosis distortion caused by the missed judgment of appearance damage, and improves the comprehensiveness and reliability of the fault preliminary judgment.

[0015] (2) the application collects the current characteristic data and current environmental data of the lightning arrester, calls the reference current characteristic data of the lightning arrester of the same type in the historical normal operation condition of the substation, corrects the reference current characteristic data based on the current environmental data, solves the problem of comparison distortion of historical and current current data caused by environmental factors, makes the comparison analysis more in line with the actual operation condition, effectively reduces the fault misjudgment rate, and improves the accuracy of the diagnosis result.

[0016] (3) The application matches the initial fault type according to the current derived parameter, and determines the specific fault type of the lightning arrester by reasonably verifying the historical environment data of each historical fault case of the same initial fault type, accurately identifies the specific fault root such as insulation aging and insulation dampening, provides a clear maintenance direction for maintenance personnel, and significantly improves the maintenance efficiency and pertinence.

[0017] (4) The application compares the current characteristic data with the historical current characteristic data of the lightning arrester itself at different historical time points in the recent period retrieved from the substation historical database, analyzes the mean deviation rate and trend change slope of the current characteristic data, and performs corresponding processing measures, so as to capture the performance degradation trend of the equipment in advance, realize early warning of faults, and reduce the outage risk of the power system caused by sudden faults of the lightning arrester. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0019] Figure 1 It is a schematic diagram of the method steps of the application.

[0020] Figure 2 It is a schematic diagram of the lightning arrester appearance damage judgment step in the application.

[0021] Figure 3 It is a schematic diagram of the current characteristic data correction step in the application.

[0022] Figure 4 It is a schematic diagram of the lightning arrester specific fault type diagnosis process step in the application. DETAILED DESCRIPTION

[0023] Various exemplary embodiments of the application will now be described in detail with reference to the accompanying drawings. Note that: unless otherwise specifically stated, the relative arrangement, numerical expressions and values of the components and steps set forth in these embodiments do not limit the scope of the application. At the same time, it should be understood that, in order to facilitate description, the sizes of the various parts shown in the drawings are not drawn in accordance with the actual proportional relationship.

[0024] The following description of at least one example embodiment is merely illustrative in nature and is in no way intended to limit the application or its application or use. Techniques, methods and devices known to those skilled in the relevant art can not be discussed in detail, but in appropriate cases, the techniques, methods and devices should be considered as part of the specification.

[0025] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary, and not as a limitation. Thus, other examples of the exemplary embodiments can have different values.

[0026] Referring to Figure 1 As shown in the figure, the application provides a smart sensor-based fault diagnosis method for substation lightning arrester monitoring, which comprises the following steps: S1, collecting the appearance image of the substation lightning arrester after being subjected to lightning action by a smart sensor, and analyzing and judging whether the lightning arrester has appearance damage based on the characteristic parameters in the appearance image.

[0027] In an embodiment of the application, before the lightning arrester fault diagnosis is performed, the appearance state of the lightning arrester after lightning is first monitored and judged, because the lightning arrester is the core equipment for resisting lightning and overvoltage in the power system, and its operating state is directly related to the stability and safety of the power system, and lightning is one of the main causes of lightning arrester failure.

[0028] Considering that the lightning arrester after lightning may not only have internal electrical performance degradation problems, but also have appearance damage types such as shell damage, insulation layer cracking and umbrella skirt deformation, if such appearance damage is not discovered in time, on the one hand, it will lead to distorted subsequent electrical performance monitoring data, affecting the accuracy of fault diagnosis, on the other hand, it may cause structural failure to expand, further threatening the safe operation of the equipment, so the appearance state of the lightning arrester needs to be judged first at the initial stage of fault diagnosis.

[0029] Based on this, in a preferred embodiment of the application, as Figure 2 shown, the basis for judging whether the lightning arrester has appearance damage is as follows: S11, collecting the appearance image of the substation lightning arrester after being subjected to lightning action by a visual sensor, and extracting the characteristic parameters of the appearance image by image recognition technology. The characteristic parameters are key image parameters that can reflect whether the appearance of the lightning arrester is damaged or cracked.

[0030] S12, matching the extracted characteristic parameters with the damage characteristic parameters of each appearance damage type one by one.

[0031] S13, when the characteristic parameters match the damage characteristic parameters of a certain appearance damage type, it indicates that the lightning arrester has this type of appearance damage, then it is determined that the lightning arrester has a structural failure, and a diagnosis report for replacement is generated, so that replacement measures can be taken in time to avoid the expansion of the failure.

[0032] It should be noted that the image acquisition process of the visual sensor and the process of extracting the characteristic parameters by image recognition technology are both prior art means, and will not be described in detail.

[0033] The damage characteristic parameters of each preset appearance damage type are obtained based on statistical analysis of a large number of image data of the same type of lightning arrester appearance damage cases, and can accurately reflect the core characteristics of different appearance damages. The implementer can fine-tune according to the model, material and other factors of the lightning arrester in the actual application scene.

[0034] The application realizes the cooperative monitoring of the appearance state and the electrical performance by collecting the appearance image of the lightning arrester after being subjected to the action of lightning in the substation, analyzing and judging whether the lightning arrester has appearance damage based on the characteristic parameters in the appearance image, avoiding the distortion of subsequent diagnosis caused by missed judgment of appearance damage, covering both the electrical performance dimension and the structural integrity dimension in the initial fault judgment, and improving the comprehensiveness and reliability of the initial fault judgment, thereby laying a solid foundation for further carrying out accurate fault diagnosis and ensuring stable operation of the power system.

[0035] S2, if the lightning arrester does not have appearance damage, the current characteristic data and the current environment data of the lightning arrester are collected synchronously, wherein the current characteristic data includes total leakage current and resistive current.

[0036] It is considered that when the lightning arrester has no appearance damage, the fault risk mainly comes from the internal electrical performance degradation, such as insulation aging, insulation dampening, etc., and the total leakage current and the resistive current are the core indexes directly reflecting the internal electrical state.

[0037] Based on this, the total leakage current is the total current flowing through the lightning arrester under operating voltage, and the abnormal increase of the value may mean that the internal insulation has a overall degradation trend. The resistive current is a component of the total leakage current, and insulation aging, dampening and other faults will cause the resistive current proportion to rise, which is the key basis for judging the internal partial defects of the lightning arrester.

[0038] It is also considered that the current characteristic data is easily affected by the environment temperature and the environment humidity, for example: high temperature may cause the internal medium conductance of the lightning arrester to increase, so that the total leakage current slightly rises; high humidity may reduce the surface resistivity of the insulation, and interfere with the true value of the resistive current. If only the current characteristic data is collected and the environmental factors are ignored, when compared with the reference current data of the historical normal working conditions, the current fluctuation caused by the environmental difference will be misjudged as a fault. Therefore, the current environment data is collected synchronously to ensure that the comparison benchmark of the historical data and the current data is consistent, which significantly reduces the fault misjudgment rate and improves the accuracy of the diagnosis result.

[0039] S3, the reference current characteristic data of the same type of lightning arrester in the historical normal operation condition of the substation is called, and the reference current characteristic data is corrected based on the current environment data.

[0040] Considering that the essence of the arrester fault diagnosis is to judge whether the current characteristic data deviates from the normal range, if there is a lack of stable and fitted normal operation current reference, it will not be able to effectively distinguish whether the current data fluctuation is a normal characteristic of the equipment or a fault signal, resulting in that the subsequent fault judgment loses basis.

[0041] In addition, considering that different models and specifications of arrester have large differences in structural design and electrical parameters, if the historical data of non-similar arrester is used as a reference, the reference will be distorted due to the mismatch of basic performance. For example, the normal leakage current range of a certain model of arrester may be significantly different from another model, and cross-type reference will directly affect the accuracy of fault judgment.

[0042] Therefore, in a specific embodiment, the current characteristic data of the same type of arrester in the historical normal operation condition in the substation is retrieved, and the specific method is as follows: first, the current characteristic data of the same type of arrester in each historical normal operation condition within a set historical period is retrieved from the historical database of the substation.

[0043] Then, the abnormal values of the total leakage current and the resistive current in each historical normal operation condition are identified and removed.

[0044] Finally, the mean values of the total leakage current and the resistive current in each historical normal operation condition after removal are calculated, and the mean values of the total leakage current and the resistive current are taken as the reference current characteristic data.

[0045] Considering that the historical data of the same type of arrester stored in the historical database of the substation may include abnormal current values caused by temporary interference, if the untreated data is directly used to calculate the reference, the reference current mean value will be too high or too low, and then the normal current will be misjudged as a fault or the fault current will be misjudged as normal, which is also a key problem of the lack of reference reliability in the prior art.

[0046] Therefore, the present application retrieves the normal condition current data of the same type of arrester within a set historical period from the historical database of the substation, and realizes the identification of abnormal values through The specific method is to calculate the mean value and the standard deviation, remove the abnormal points exceeding the mean value ± 3 times the standard deviation, and calculate the mean values of the total leakage current and the resistive current after removing the interference data, and finally determine the reference current characteristic data, so as to ensure that the reference is highly fitted with the normal operation state of the current arrester.

[0047] Considering that the environmental conditions of the historical normal operation condition and the current environmental conditions may have large differences, if the environmental adaptation is not performed, the comparison reference of the reference data and the current data will be inconsistent, and the electrical performance state of the current arrester cannot be truly reflected.

[0048] Therefore, the application screens historical environment data with the highest similarity to the reference current characteristic data as reference environment data, counts current and environment data of other historical normal operating conditions, fits environment correction equations of total leakage current and resistive current, substitutes the current environment data into the correction equations to calculate correction coefficients, and finally corrects the reference current characteristic data into equivalent data consistent with the current environment operating condition, so as to ensure that the comparison benchmark matches the actual operating condition.

[0049] Preferably, as shown in the embodiment of the application, Figure 3 The correction of the reference current characteristic data is specifically as follows: A1, screening historical normal operating conditions with the highest similarity to the reference current characteristic data from historical normal operating conditions, and taking historical environment data of the historical normal operating conditions as reference environment data.

[0050] The historical normal operating condition with the highest similarity refers to a historical normal operating condition with the minimum deviation of historical current characteristic data from the reference current characteristic data.

[0051] A2, screening other historical normal operating conditions corresponding to the reference environment data, and counting current characteristic data and historical environment data of each other historical normal operating condition.

[0052] A3, performing ratio analysis on the current characteristic data and the reference environment data to obtain current characteristic correction coefficients and environment data ratios.

[0053] A4, substituting the environment data ratios into the multiple linear regression equation as independent variables, and substituting total leakage current correction coefficients and resistive current correction coefficients in the current characteristic correction coefficients into the multiple linear regression equation as dependent variables, and fitting the total leakage current correction equation and the resistive current correction equation by using the least square method.

[0054] A5, substituting the current environment data into the fitted correction equation to output corresponding total leakage current correction coefficients and resistive current correction coefficients.

[0055] A6, multiplying the reference current characteristic data with the total leakage current correction coefficients and the resistive current correction coefficients to obtain corrected total leakage current and resistive current.

[0056] The application collects current characteristic data and current environment data of the lightning arrester, retrieves reference current characteristic data of the same type of lightning arrester in the historical normal operating condition in the substation, corrects the reference current characteristic data based on the current environment data, solves the problem of comparison distortion of historical and current current data caused by environmental factors, makes the comparison analysis more consistent with the actual operating condition, effectively reduces the fault misjudgment rate, and improves the accuracy of the diagnosis result.

[0057] S4, compare the collected current characteristic data with the corrected reference current characteristic data, and judge whether the lightning arrester is in a fault state.

[0058] Considering that the current data of the lightning arrester will exist small natural fluctuations in normal operation, if the allowable tolerance range is not set, the normal fluctuations will be misjudged as faults, resulting in unnecessary increase of maintenance cost.

[0059] Based on this, in a specific embodiment, the judgment of whether the lightning arrester is in a fault state specifically includes: first, difference comparison is performed between the collected current characteristic data and the corrected reference current characteristic data, to obtain full leakage current deviation value and resistive current deviation value respectively.

[0060] Then, if the full leakage current deviation value or the resistive current deviation value exceeds the allowable tolerance range of the corresponding characteristic, it is judged that the lightning arrester is in a fault state, otherwise, it is judged that the lightning arrester is not in a fault state.

[0061] The present application judges whether the lightning arrester is in a fault state by calculating the full leakage current deviation value and the resistive current deviation value, and comparing them with the corresponding allowable tolerance range, provides clear basis for subsequent diagnosis of whether to enter specific fault type, simplifies the diagnosis process under normal working conditions, and improves the overall diagnosis efficiency.

[0062] S5, when the lightning arrester is in a fault state, comprehensive comparison and analysis are performed in combination with the collected current characteristic data and the current environmental data, to diagnose the specific fault type of the lightning arrester.

[0063] Considering that the maintenance measures of different fault types are significantly different, such as replacement of insulation parts for insulation aging and drying treatment for insulation damp, if only the existence of faults is known but the fault type is not known, the maintenance time will be prolonged and the maintenance cost will be increased.

[0064] Also considering that single current characteristic data cannot distinguish fault types, for example, insulation aging and insulation damp will both cause resistive current to rise, but insulation damp will be accompanied by higher abnormal increase of full leakage current in high humidity environment, and insulation aging will have more obvious resistive current increment in high temperature environment, so current derivative parameters such as resistive current proportion, temperature resistive current increment, humidity full leakage current abnormal increment, etc. need to be further distinguished in combination with environmental data.

[0065] It is also considered that the current performance of the same fault type may be different under different environments, if the fault type is only initially judged by matching the current derivative parameters without combining with environmental reasonableness verification, the fault type will be misjudged due to environmental mismatch.

[0066] Based on this, as Figure 4As shown, the specific fault type of the diagnostic lightning arrester is as follows: S51, based on the collected current characteristic data and current environmental data, the resistive current proportion, the resistive current increment under the current environmental temperature, and the abnormal increase of the total leakage current under the current environmental humidity are determined as the current derivative parameters.

[0067] Among them, the resistive current proportion is the proportion of resistive current in total leakage circuit.

[0068] The resistive current increment under the current environmental temperature is the difference between the collected resistive current and the corrected resistive current. This difference is a core quantitative index for measuring the abnormality of the internal insulation state of the lightning arrester. Its essence is to reflect the abnormal growth degree of the resistive current by eliminating the deviation of the actual resistive current after temperature interference from the normal reference value.

[0069] The abnormal increase of the total leakage current under the current environmental humidity is the ratio of the collected total leakage current to the corrected total leakage current. This ratio reflects the humidity adaptability deviation of the actual current relative to the normal state. When there is no fault, the ratio should be close to 1, and internal dampness and other faults will cause the ratio to be significantly greater than 1.

[0070] S52, match the current derivative parameters with the set current derivative parameter threshold values of each fault type, and take the matched fault type as the preliminary fault type.

[0071] It should be noted that when all parameters in the current derivative parameters are higher than the current derivative parameter threshold value of a certain fault type, the fault type is taken as the matched fault type.

[0072] S53, retrieve the fault types of different historical fault cases of the same type lightning arrester from the substation historical database, and screen each historical fault case with the same preliminary fault type.

[0073] S54, statistics the historical environmental data of each screened historical fault case, and reasonably verifies the current environmental data with the historical environmental data of each historical fault case to determine the specific fault type of the diagnostic lightning arrester.

[0074] Among them, the setting of the current derivative parameter threshold value of each fault type needs to be based on the statistics of historical fault cases of the same type lightning arrester. Considering that the current derivative parameters of different fault types have minimum threshold values, for example, the resistive current proportion of insulation aging needs to reach 30% to be determined as this fault, otherwise it does not belong to it, and the minimum value of each fault type corresponding to the current derivative parameter is selected as the threshold value, which can avoid misjudging the slight abnormality that does not reach the fault degree as the corresponding fault type, and ensure the scientificity of the threshold value setting.

[0075] As an example, the current derivative parameter threshold setting mode corresponding to each fault type is: first, according to the fault type of different historical fault cases of the same type of lightning arrester, the current derivative parameter corresponding to each fault type in different historical fault cases is counted.

[0076] Then, based on the current characteristic data and historical environment data in the different historical fault cases corresponding to each fault type, the current derivative parameter in the different historical fault cases corresponding to each fault type is obtained.

[0077] Finally, the minimum value of the current derivative parameter corresponding to each fault type is compared and screened as the corresponding current derivative parameter threshold.

[0078] Preferably, in an embodiment of the present application, the rationality verification of the current environment data and the historical environment data of each historical fault case comprises: S541, determining the maximum value and the minimum value of the historical environment data according to the historical environment data of the screened historical fault cases.

[0079] S542, based on the historical environment data range formed by the maximum value and the minimum value of the historical environment data, comparing and verifying the current environment data with the historical environment data range, if the current environment data is within the historical environment data range, the current environment data is verified to be reasonable, and the preliminary fault type is taken as the specific fault type.

[0080] S543, otherwise, the preliminary fault type is re-determined until the current environment data is verified to be reasonable, and the preliminary fault type is taken as the specific fault type.

[0081] The present application matches the preliminary fault type according to the current derivative parameter, verifies the rationality through the historical environment data of the same historical fault case as the initial fault type, determines the specific fault type of the lightning arrester, accurately identifies the specific fault root such as insulation aging and insulation damp, provides a clear maintenance direction for maintenance personnel, and significantly improves the maintenance efficiency and pertinence.

[0082] S6, otherwise, the current characteristic data is compared with the historical current characteristic data corresponding to the lightning arrester, and the corresponding processing measures are executed based on the longitudinal comparison result.

[0083] Considering that lightning arrester failures also include chronic deterioration failures, the current characteristic change of this type of failure has a gradual nature, and when comparing the current data with the historical normal reference data of the same type of lightning arrester at a time, the current data may be in the allowed normal range, and the historical current characteristic data of the lightning arrester is a cumulative record of the real running state, and longitudinal comparison can exclude individual difference interference and more accurately capture exclusive abnormal signals.

[0084] Based on this, in a specific embodiment, the longitudinal comparison of the current characteristic data with the historical current characteristic data corresponding to the surge arrester includes: S61, comparing the current characteristic data with the historical current characteristic data of the surge arrester at different historical time points in the recent period retrieved from the substation historical database, and analyzing the mean deviation rate and the trend change slope of the current characteristic data.

[0085] S62, if the mean deviation rate of the current characteristic data exceeds the set normal fluctuation threshold and the trend change slope shows an increasing trend, a state warning signal is generated, otherwise the surge arrester is in a normal operating state.

[0086] Wherein, considering that when the surge arrester is in normal operation, the current data will naturally fluctuate in a small range, and only the deviation of the current data from the historical mean value cannot distinguish between normal fluctuation and abnormal degradation, a normal fluctuation threshold is set, and the difference exceeding the threshold indicates that the difference has exceeded the natural range.

[0087] It is also considered that the deviation exceeding the threshold may be a single accidental fluctuation, which needs to be verified by the trend change slope whether there is a continuous deterioration trend, for example, the slope is positive and increasing, which means that the current is rising at an accelerating rate, and the combination of the two can avoid misjudging accidental fluctuation as abnormal or missing continuous degradation, ensuring the rigor of the judgment logic.

[0088] Based on this, first, the deviation value of the current characteristic data from the mean value of the historical current characteristic data at different historical time points in the recent period is obtained, and then the ratio of the mean value of the historical current characteristic data at different historical time points in the recent period to the current characteristic data is taken as the mean deviation rate.

[0089] First, the historical current characteristic data at different historical time points in the recent period and the collected current characteristic data are sorted by time, then the linear regression is used to fit the change curve of the current characteristic data with time, and finally the trend change slope is obtained based on the change curve.

[0090] It should be noted that the recent period of the surge arrester itself can be the recent month or the recent half month, and by retrieving the historical current characteristic data at different historical time points in the recent period, the current performance baseline of the device can be better reflected, ensuring the reference of the longitudinal comparison.

[0091] The above processes of retrieving historical data, fitting curves by linear regression, etc. are all prior art means and will not be described in detail.

[0092] The present application compares the current characteristic data with the historical current characteristic data of the surge arrester at different historical time points in the recent period retrieved from the substation historical database, analyzes the mean deviation rate and the trend change slope of the current characteristic data, and performs corresponding processing measures, so as to capture the performance degradation trend of the device in advance, realize early warning of failure, and reduce the risk of outage of the power system caused by sudden failure of the surge arrester.

[0093] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product.

[0094] Those skilled in the art can realize that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0095] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0096] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0097] Finally, the above is merely preferred embodiments of the present application, and is not intended to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors, characterized in that, include: Collect visual images of the surge arresters in the substation after they have been struck by lightning to characterize the feature parameters and determine whether there is any visual damage to the surge arresters. If there is no visible damage, the current characteristic data of the surge arrester and the current environmental data are collected simultaneously. The current characteristic data includes the total leakage current and resistive current. Substitute the current environmental data into the fitted total leakage current and resistive current correction equations, output the total leakage current correction coefficient and resistive current correction coefficient, and correct the reference current characteristic data of the same type of surge arrester in the substation. The current characteristic data is compared and analyzed with the corrected reference current characteristic data to determine whether the surge arrester is in a fault state. When the system is in a fault state, the current-derived parameters are determined by combining current characteristic data and current environmental data. An initial fault type is then identified, and its validity is verified against historical fault cases of the same type to determine the specific fault type of the surge arrester. The details are as follows: Based on the collected current characteristic data and current environmental data, the proportion of resistive current, the increase of resistive current at the current ambient temperature, and the abnormal increase of total leakage current at the current ambient humidity are determined and used as current-derived parameters. The current-derived parameters are matched with the current-derived parameter thresholds corresponding to each fault type, and the fault types that are successfully matched are used as the initial fault types. Retrieve the fault types of different historical fault cases of the same type of surge arrester from the substation historical database, and filter out the historical fault cases that are the same as the initial fault type. The historical environmental data of each historical fault case is statistically analyzed, and the current environmental data is verified to be reasonable with the historical environmental data of each historical fault case to determine the specific fault type of the surge arrester. Specifically, this includes: determining the maximum and minimum values ​​of historical environmental data based on the historical environmental data of each screened historical failure case; The historical environmental data range is defined by the maximum and minimum values ​​of historical environmental data. The current environmental data is compared and verified with the historical environmental data range. If the current environmental data is within the historical environmental data range, the rationality verification of the current environmental data is passed, and the initial fault type is adopted as the specific fault type. Otherwise, the initial fault type is redefined until the rationality verification of the current environmental data is passed, and the initial fault type is adopted as the specific fault type. When the system is in a non-faulty state, the current characteristic data is compared with the recent historical current characteristic data of the surge arrester to determine the mean deviation rate and the slope of the trend change, and corresponding processing measures are implemented accordingly.

2. The method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors according to claim 1, characterized in that: The criteria for determining whether a surge arrester has external damage are as follows: Visual sensors are used to collect images of the appearance of the lightning arresters in the substation after they have been struck by lightning. Image recognition technology is then used to extract the characteristic parameters of the appearance images. The extracted characterization parameters are matched one by one with the preset damage characteristic parameters for each type of appearance damage. When the characterization parameter matches the damage characteristic parameter of a certain type of appearance damage, it indicates that the surge arrester has that type of appearance damage. Therefore, it is determined that the surge arrester has a structural fault, and a diagnostic report indicating that it needs to be replaced is generated.

3. The method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors according to claim 1, characterized in that: The method for retrieving the reference current characteristic data of the same type of surge arrester in the substation is as follows: Retrieve current characteristic data of the same type of surge arrester under various historical normal operating conditions within a set historical time period from the substation historical database; Outliers in the total leakage current and resistive current under each historical normal operating condition are identified and eliminated. The average values ​​of the total leakage current and resistive current under each historical normal operating condition after removal are calculated, and the average values ​​of the total leakage current and resistive current are used as reference current characteristic data.

4. The method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors according to claim 3, characterized in that: The method for correcting the reference current characteristic data of similar surge arresters in substations is as follows: Select the historical normal operating condition that has the highest similarity to the reference current characteristic data from each historical normal operating condition, and use the historical environmental data under this historical normal operating condition as the reference environmental data; Filter out other historical normal operating conditions besides the historical normal operating conditions corresponding to the reference environmental data, and statistically analyze the current characteristic data and historical environmental data under each other historical normal operating condition; By comparing these values ​​with reference current characteristic data and reference environmental data respectively, the current characteristic correction coefficient and the ratio of environmental data are obtained. Using the ratio of environmental data as the independent variable, and the total leakage current correction coefficient and the resistive current correction coefficient in the current characteristic correction coefficient as the dependent variable, we substituted them into the multiple linear regression equation and used the least squares method to fit and obtain the total leakage current correction equation and the resistive current correction equation. Substitute the current environmental data into the fitted correction equations to output the corresponding total leakage current correction coefficient and resistive current correction coefficient. The reference current characteristic data are multiplied by the total leakage current correction factor and the resistive current correction factor, respectively, to obtain the corrected total leakage current and resistive current.

5. The method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors according to claim 1, characterized in that: The determination of whether the surge arrester is in a fault state specifically includes: The difference between the collected current characteristic data and the corrected reference current characteristic data is compared to obtain the total leakage current deviation value and the resistive current deviation value, respectively. If the total leakage current deviation or resistive current deviation exceeds the allowable tolerance range of the corresponding characteristic, the surge arrester is judged to be in a fault state; otherwise, the surge arrester is judged not to be in a fault state.

6. The method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors according to claim 1, characterized in that: The threshold setting method for the current-derived parameters corresponding to each fault type is as follows: Based on the different historical fault cases of the same type of surge arrester, statistics were compiled on the different historical fault cases corresponding to each fault type. Based on the current characteristic data and historical environmental data of different historical fault cases corresponding to each fault type, the current derived parameters of each fault type corresponding to different historical fault cases are obtained. The minimum value of the current-derived parameter corresponding to each fault type is compared and selected as the threshold value of the corresponding current-derived parameter.

7. The method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors according to claim 1, characterized in that: The comparison of current characteristic data with the surge arrester's own recent historical current characteristic data specifically includes: The current characteristic data is compared with the historical current characteristic data of the surge arrester itself at different historical time points in recent times, retrieved from the historical database of the substation, and the mean deviation rate and trend change slope of the current characteristic data are analyzed. If the mean deviation rate of the current characteristic data exceeds the set normal fluctuation threshold and its trend slope is increasing, a status warning signal is generated; otherwise, the surge arrester is in normal operating condition.

8. A method for monitoring and diagnosing faults in substation surge arresters based on intelligent sensors according to claim 7, characterized in that: The analysis method for the mean deviation rate and trend change slope of the current characteristic data is as follows: Obtain the deviation value between the current characteristic data and the mean value of historical current characteristic data at different recent historical time points, and use the ratio of the deviation value to the mean value of historical current characteristic data at different recent historical time points as the mean deviation rate. Historical current characteristic data from different recent historical time points and collected current characteristic data are sorted by time. Linear regression is used to fit the change curve of current characteristic data over time, and the slope of trend change is obtained based on the change curve.

Images