Overvoltage inversion system and method based on lightning arrester tail end current monitoring

Abstract

The invention relates to the technical field of lightning arrester overvoltage on-line monitoring, in particular to an overvoltage inversion system and method based on lightning arrester tail end current monitoring, and the system comprises a pressurizer, a lightning collection module, a current collection module, an intelligent data analysis end, an inversion relation verification module and an inversion relation new building module, the lightning simulation data are applied to the lightning arrester, the lightning real data are collected, the current collection module is installed at the tail end of the lightning arrester and collects tail end current simulation data and tail end current real data, and the intelligent data analysis end establishes and stores the corresponding relation between the lightning simulation data and the tail end current simulation data. Using an inversion relation verification module to analyze whether the tail end current real data has corresponding lightning real data, and if not, using an inversion relation new module to establish a correct overvoltage inversion relation; according to the invention, continuous creation of the inversion relationship is realized through the inversion relationship verification module and the new module, and the machine learning module can continuously construct a high-precision inversion relationship based on feedback information of the manual construction module, so that the precision of the inversion relationship is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of arrester overvoltage online monitoring, and in particular to an overvoltage inversion system and method based on arrester tail current monitoring. Background Art

[0002] Overvoltage refers to the phenomenon of long-term voltage fluctuation when the root mean square value of AC voltage under power frequency rises, exceeds 10% of the rated value, and lasts for more than 1 minute. Among them, external overvoltage, that is, lightning overvoltage, is difficult to monitor during the operation of the power grid, especially in high-altitude areas. Lightning will have a profound impact on the power system. Under the influence of lightning, the current at the tail end of the lightning arrester may change, and this change may cause damage to power equipment. Therefore, it is necessary to confirm the performance of the lightning arrester based on the voltage inversion system. At present, some overvoltage inversion systems based on lightning arrester tail current monitoring have been developed, such as patent application CN115267305A, which realizes overvoltage inversion by using simulated lightning signals based on lightning arrester tail current monitoring, that is, it is equivalent to directly equating the lightning value with the overvoltage value, which is also the main method of lightning overvoltage inversion monitoring at present.

[0003] The current overvoltage inversion system for monitoring the current at the tail end of the arrester has the following problems: (1) The error accumulation is too large. In the overvoltage inversion process, the main method currently used is to achieve overvoltage inversion by setting the simulation signal of lightning. However, this simulation signal obviously cannot completely simulate the lightning data. In multiple overvoltage inversion operations, due to the deviation of the set simulation signal, the error of the overvoltage inversion result obtained will also accumulate, and the inversion result obtained by the entire system is difficult to accurately reflect the actual situation; (2) The state of the arrester cannot be monitored. In the overvoltage inversion system for monitoring the current at the tail end of the arrester that is currently used, the main method is to directly use the inversion system after it is constructed, and at the same time, other arrester status monitoring systems are constructed to analyze the operating status of the arrester. This will make the complexity of the entire system too high and the failure rate will increase. Summary of the invention

[0004] In order to solve the problems of excessive error accumulation of overvoltage inversion system and inability to monitor the state of lightning arrester in the prior art, the present application discloses the following technical solutions:

[0005] An overvoltage inversion system based on arrester tail current monitoring, comprising:

[0006] A pressurizing end, mounted on the top of the arrester, for applying lightning simulation data to the arrester;

[0007] A lightning collection module is installed on the top of the lightning arrester and is used to collect real lightning data;

[0008] A current acquisition module is installed at the tail end of the arrester, and is used to collect the tail end current simulation data of the arrester and the tail end current real data of the arrester; the tail end current simulation data is the current data generated under the action of the lightning simulation data, and the tail end current real data is the current data generated under the action of the lightning real data;

[0009] An intelligent data analysis terminal, connected to the current acquisition module, the lightning acquisition module and the pressurizing terminal respectively, for establishing and storing a correspondence between the tail end current simulation data and the lightning simulation data and storing a correspondence between the tail end current real data and the lightning real data;

[0010] An inversion relationship verification module, connected to the intelligent data analysis terminal, for performing inversion and verification using the corresponding relationship between the tail end current simulation data and the lightning simulation data;

[0011] The inversion relationship creation module is connected to the intelligent data analysis terminal and is used to establish the corresponding relationship between the real lightning data and the real tail current data after inversion and verification.

[0012] Optionally, the intelligent data analysis terminal includes:

[0013] A lightning simulation data acquisition module, connected to the pressurizing end, for acquiring the lightning simulation data;

[0014] A tail end current simulation data acquisition module, connected to the current acquisition module, for acquiring the tail end current simulation data;

[0015] An inversion relationship creation module, connected to the lightning simulation data acquisition module and the tail current simulation data acquisition module respectively, for establishing a corresponding relationship between the lightning simulation data and the tail current simulation data;

[0016] The inversion relationship storage module is connected to the inversion relationship creation module and the inversion relationship verification module, and is used to store the corresponding relationship between the tail end current simulation data and the lightning simulation data and the corresponding relationship between the tail end current real data and the lightning real data.

[0017] Optionally, the inversion relationship verification module includes:

[0018] A tail end current real data acquisition unit, connected to the current acquisition module, for acquiring the tail end current real data;

[0019] A lightning real data acquisition unit, connected to the lightning acquisition module, for acquiring the lightning real data;

[0020] a first judgment unit connected to the tail current real data acquisition unit, and used to judge whether the tail current real data exists in the intelligent data analysis terminal; if it exists in the intelligent data analysis terminal, the lightning simulation data is inverted according to the corresponding relationship between the lightning simulation data and the tail current simulation data to obtain the lightning simulation inversion data;

[0021] a deviation value calculation unit, connected to the lightning real data acquisition unit and the first judgment unit respectively, and used to calculate the deviation value between the lightning simulation inversion data and the lightning real data;

[0022] The second judgment unit is connected to the deviation value calculation unit, and is used to judge whether the deviation value is within the deviation value threshold range. If it is within the deviation value threshold range, it means that the correspondence between the real data of the tail end current and the lightning simulation data has passed the verification, and the verification result is sent to the intelligent data analysis end for storage; if it is not within the deviation value threshold range, it means that the correspondence between the real data of the tail end current and the lightning simulation data has not passed the verification.

[0023] Optionally, the system further includes:

[0024] A machine learning module is connected to the inversion relationship verification module, and is used to establish a corresponding relationship between the real data of the tail current and the real data of lightning when the corresponding relationship between the real data of the tail current and the lightning simulation data fails to pass the verification, so as to form a temporary overvoltage inversion relationship;

[0025] An evaluation module, connected to the machine learning module, for determining whether the temporary overvoltage inversion relationship is correct; if the temporary overvoltage inversion relationship is correct, the temporary overvoltage inversion relationship is authenticated as a new overvoltage inversion relationship; if the temporary overvoltage inversion relationship is wrong, the temporary overvoltage inversion relationship is deleted;

[0026] The manual construction module is connected to the machine learning module and is used for, when the temporary overvoltage inversion relationship is wrong, for online experts to intervene, to construct a new overvoltage inversion relationship, and to send the construction results and influencing factors to the machine learning module.

[0027] Optionally, the influencing factors include arrester status data, arrester operation time, number of lightning strikes sustained by the arrester, arrester model, number of arresters connected in series, meteorological conditions in the area where the arrester is located, and arrester initial status data.

[0028] An overvoltage inversion method based on arrester tail current monitoring includes the following steps:

[0029] Apply lightning simulation data to the arrester using the pressurized end;

[0030] Use the lightning collection module installed on the top of the lightning arrester to collect real lightning data;

[0031] The current collection module installed at the tail end of the arrester is used to collect the tail end current simulation data of the arrester and the tail end current real data of the arrester; the tail end current simulation data is the current data generated under the action of the lightning simulation data, and the tail end current real data is the current data generated under the action of the lightning real data;

[0032] Using an intelligent data analysis terminal to establish and store a correspondence between the tail end current simulation data and the lightning simulation data, and to store a correspondence between the tail end current real data and the lightning real data;

[0033] Using an inversion relationship verification module to perform inversion and verification using the corresponding relationship between the tail current simulation data and the lightning simulation data;

[0034] The corresponding relationship between the real lightning data and the real tail current data is established by using the inversion relationship new creation module after the inversion and verification are passed.

[0035] Optionally, the inversion relationship verification module uses the corresponding relationship between the tail current simulation data and the lightning simulation data to perform inversion and verification, specifically including:

[0036] Determine whether the tail end current real data exists in the intelligent data analysis terminal; if it exists in the intelligent data analysis terminal, invert the lightning simulation data according to the corresponding relationship between the lightning simulation data and the tail end current simulation data to obtain lightning simulation inversion data;

[0037] Calculating the deviation between the lightning simulation inversion data and the lightning real data;

[0038] Determine whether the deviation value is within the deviation value threshold range. If so, it means that the correspondence between the real data of the tail end current and the lightning simulation data has been verified, and the verification result is sent to the intelligent data analysis end for storage; if not, it means that the correspondence between the real data of the tail end current and the lightning simulation data has not been verified.

[0039] The beneficial effects of this application are:

[0040] 1. The maximum processing of error accumulation is achieved. In the technical solution of this application, in addition to constructing the correspondence between the tail current simulation data and the lightning simulation data as a basic lightning data inversion relationship based on the tail current data, the voltage inversion data in the real situation is collected based on the machine learning module, and based on this type of collected real lightning data, the inversion system can be continuously updated to fully avoid error accumulation.

[0041] 2. Realize arrester status monitoring. Based on the overvoltage inversion system, the actual operation data of the arrester is directly obtained, including lightning simulation data, tail current simulation data, tail current real data and lightning real data. When the tail current simulation data and the tail current real data are the same, the corresponding lightning simulation data and lightning real data deviation value are calculated. If the data deviation value exceeds the set threshold, it is confirmed that the current transformer has a fault, thereby realizing the status monitoring of the transformer.

[0042] 3. Realize intelligent monitoring. For the overvoltage inversion system of transformer current monitoring, if the simulation is based solely on preset simulation data, some overvoltage data that have not been verified will be difficult to invert. Based on the data collection and corresponding relationship confirmation of the machine learning module, combined with the data analysis of the manual construction module, the machine learning module can continuously complete the overvoltage inversion based on the data analysis process of the manual construction module to ensure the accuracy of the overvoltage inversion. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solution of the exemplary embodiment of the present invention, the following briefly introduces the drawings required to describe the embodiment. Obviously, the drawings introduced are only some of the embodiment drawings to be described in the present invention, rather than all the drawings. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 A schematic diagram of the structure of an overvoltage inversion system based on arrester tail current monitoring provided by an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of the operation flow of an intelligent data analysis terminal of an overvoltage inversion system based on arrester tail current monitoring provided by an embodiment of the present invention;

[0046] Figure 3 An operation flow chart of an overvoltage inversion system based on arrester tail current monitoring provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0048] It should also be noted that, for ease of description, only the part relevant to the present invention but not all content is shown in the accompanying drawings. It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processing or methods depicted as flow charts. Although the flow chart describes each operation (or step) as sequential processing, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. When its operation is completed, the processing can be terminated, but it can also have additional steps not included in the accompanying drawings. The processing can correspond to methods, functions, procedures, subroutines, subprograms, etc.

[0049] Embodiment 1

[0050] In the construction of the overvoltage inversion system based on current monitoring, the method adopted by the present invention is also to construct a correspondence between the monitored current and the lightning voltage. However, in the construction of the specific correspondence, other structures are set up to solve the problems existing in the prior art.

[0051] Figure 1 It is a structural schematic diagram of an overvoltage inversion system based on current monitoring at the tail end of a lightning arrester provided in an embodiment of the present invention. This embodiment is suitable for overvoltage inversion operations on current monitoring at the tail end of a lightning arrester, wherein an intelligent data analysis terminal obtains a variety of data, including lightning simulation data at the head end of the lightning arrester, real lightning data at the head end, and simulation and real data of current at the tail end of the lightning arrester.

[0052] like Figure 1 As shown, the overvoltage inversion system structure based on the current monitoring of the tail end of the arrester of the present invention is composed of an arrester and an intelligent data analysis terminal, wherein the arrester is set according to the actual situation, and the data of the arrester is recorded based on the model, resistance value, use time and other data of the arrester. The pressurized end of the arrester is the lightning simulation data input terminal, that is: the device setting terminal for imitating lightning data, and the pressurized end and the intelligent data analysis terminal are communicatively connected, so that the intelligent data analysis terminal can obtain the lightning simulation data applied by the pressurized end. At the same time, a lightning acquisition module is also provided, and the lightning acquisition module collects the real lightning data at the top of the arrester when the arrester is actually running. In addition to being connected to the grounding down conductor, the tail end of the arrester is also provided with a current acquisition module, which is communicatively connected to the intelligent data analysis terminal, and the intelligent data analysis terminal is communicatively connected to the lightning acquisition module and the current acquisition module.

[0053] Among them, the lightning arrester is also provided with a lightning collection module for obtaining the real lightning data that can be collected when the lightning arrester is put into use.

[0054] Among them, the set lightning arresters are tested separately according to the application time, and the lightning simulation data of the pressurized end and the measurement signal obtained by the current acquisition module are obtained at the same time. Specifically, the lightning arrester can be selected according to the application time of the lightning arrester. For example, the expected life of a certain type of lightning arrester is 8 years, but during the application process, its performance change does not show a linear relationship. At this time, the lightning arrester is screened according to the same step length, and the tail current data is recorded on the basis of the lightning simulation data, such as screening according to the scheme in the following table:

[0055] Table 1 Correspondence table of selected time and tail current of a certain type of arrester

[0056]

[0057]

[0058] All available data are recorded. It should be noted that even if the lightning simulation data is continuous, in actual operation, the tail current simulation data generated is discrete. Therefore, based on the generated inversion relationship, the real tail current data obtained in the actual situation may not have corresponding lightning simulation data. Such real tail current data and overvoltage inversion relationship need to be constructed and filled.

[0059] In some embodiments, the time step is set in other forms, such as setting the time step to half a year, for data selection.

[0060] In some embodiments, considering that the number of lightning strikes has a significant impact on the arrester, the data included therein also includes the number of overvoltages of the arrester, so as to determine the relationship between the number of overvoltages and the tail current frequency, the tail current size, the number of overvoltages and the overvoltage data, and determine the number of overvoltages based on the tail current, so as to confirm whether the arrester currently in use has failed or has other problems.

[0061] Among them, all the lightning arrester data obtained need to fully record the model, data, number of series connections and other information of the lightning arrester, and establish a corresponding relationship between the lightning arrester model and / or lightning arrester data - lightning arrester operation time - head end voltage - tail end current frequency - tail end current size. At this time, a database with the lightning arrester model as the database identifier, tail end current simulation data, and lightning simulation data can be established.

[0062] Among them, all test results for lightning arresters are transmitted to the intelligent data analysis terminal through the communication system, and the intelligent data analysis terminal constructs the correspondence between lightning simulation data and tail current simulation data, which is the overvoltage inversion relationship.

[0063] Among them, the lightning collection module is connected to the intelligent data analysis terminal, and the lightning collection module is used to obtain the real lightning data around the area where the lightning arrester is located after the lightning arrester is installed. During the operation of the lightning arrester, each time lightning occurs, the lightning collection module transmits the collected real lightning data to the intelligent data analysis terminal, obtains the inversion relationship between the real lightning data and the real data of the current at the tail end of the lightning arrester, and after confirming the new inversion relationship, the new inversion relationship is stored by the intelligent data analysis terminal.

[0064] It also includes a machine learning module, an inversion relationship verification module, an evaluation module and a manual construction module, which are used to first establish the corresponding relationship between the real data of the tail current and the real data of lightning, and obtain the temporary overvoltage inversion relationship, and then further verify the temporary overvoltage inversion relationship. This temporary overvoltage inversion relationship is verified by the evaluation module. If it fails to pass the verification, the real data of the tail current and the real data of lightning are sent to the manual construction module, and a special online expert will establish a new overvoltage inversion relationship based on all historical data of the lightning arrester, and all the generated data in the process will be fed back to the machine learning module.

[0065] The machine learning module is connected to the inversion relationship verification module, and is used to establish the corresponding relationship between the real data of the tail current and the real data of lightning when the corresponding relationship between the real data of the tail current and the lightning simulation data fails to pass the verification, so as to form a temporary overvoltage inversion relationship;

[0066] An evaluation module, connected to the machine learning module, for determining whether the temporary overvoltage inversion relationship is correct. If the temporary overvoltage inversion relationship is correct, the temporary overvoltage inversion relationship is authenticated as a new overvoltage inversion relationship. If the temporary overvoltage inversion relationship is wrong, the temporary overvoltage inversion relationship is deleted.

[0067] The manual construction module is connected to the machine learning module and is used for the intervention of online experts to construct a new overvoltage inversion relationship when the temporary overvoltage inversion relationship is wrong, and send the construction results and influencing factors to the machine learning module.

[0068] Among them, the influencing factors include the arrester status data, the arrester operation time, the number of lightning strikes the arrester has experienced, the arrester model, the number of arresters connected in series, the meteorological conditions in the area where the arrester is located, and the arrester initial status data.

[0069] Embodiment 2

[0070] Figure 2 is a schematic diagram of the operation flow of an overvoltage inversion system based on arrester tail current monitoring provided by an embodiment of the present invention, such as Figure 2As shown, this embodiment is suitable for establishing overvoltage inversion management for tail current monitoring of lightning arrester, including a tail current simulation data acquisition unit, a lightning simulation data acquisition module, an inversion relationship creation module, a tail current simulation data acquisition module, a lightning real data acquisition unit, a first judgment unit, a deviation value calculation unit, a second judgment unit, an inversion relationship storage module, a machine learning module and a manual construction module.

[0071] Among them, the intelligent data analysis end includes a tail current simulation data acquisition module, a lightning simulation data acquisition module and an inversion relationship creation module.

[0072] Among them, the tail current simulation data acquisition module, the lightning simulation data acquisition module and the inversion relationship creation module are communicatively connected, and the obtained simulation data are sent to the inversion relationship creation module to establish the corresponding relationship between the tail current simulation data and the lightning simulation data.

[0073] Among them, the inversion relationship creation module in the intelligent data analysis terminal is connected to the inversion relationship storage module to store the established corresponding relationship between the tail current simulation data and the lightning simulation data, and store the tail current simulation data and the lightning simulation data.

[0074] Among them, the inversion relationship storage module is connected to the first judgment unit, and the first judgment unit is also connected to the tail current real data acquisition module. The tail current real data acquisition module obtains the tail current real data generated by the arrester during formal operation obtained by the current acquisition module.

[0075] Among them, the current data obtained by the tail current real data acquisition module is identified by the first judgment unit to determine whether the obtained tail current real data exists in the currently constructed inversion relationship storage module, and the inversion relationship storage module is set in the intelligent data analysis terminal.

[0076] Among them, it is confirmed whether the obtained real data of the tail current has been stored in the inversion database. If so, the corresponding lightning simulation data is obtained based on the corresponding relationship. If not, it is considered that the real data of the tail current has failed to pass the verification.

[0077] Among them, the deviation value calculation unit is connected to the first judgment unit, the real lightning data is compared with the lightning simulation data obtained based on the inversion relationship, and the deviation value is calculated. If it exceeds the deviation value threshold limit range, it is considered that the correspondence between the real tail current data and the lightning simulation data is wrong, that is, the inversion relationship does not exist.

[0078] The deviation value calculation unit is connected to the second judgment unit and is used to judge whether the deviation value is within the deviation value threshold range. If not, it means that the current inversion relationship is wrong.

[0079] Among them, the machine learning module is connected to the second judgment unit to directly build a corresponding relationship between the real data of the tail current and the real data of lightning, so as to analyze the inversion relationship that can be constructed based on the real data of lightning directly collected.

[0080] Among them, the machine learning module and the manual construction module are connected in a two-way communication.

[0081] Among them, when the real data of the tail current fails to pass the verification, and the deviation value between the lightning simulation data and the real lightning data exceeds the threshold limit, all the data are sent to the machine learning module. The machine learning module directly establishes the correspondence between the real lightning data and the real data of the tail current, and sends the correspondence to the manual construction module. After analysis, the manual construction module constructs the correct inversion relationship, which is resent to the machine learning module, so that the machine learning module can achieve better data analysis and learning based on the data processing method of the manual construction module, the correction results of the algorithm, etc., so that the machine learning module can gradually imitate the ideas of the manual construction module to complete the inversion relationship construction work.

[0082] Among them, in the manual construction module, it is connected with the machine learning module, so that when the temporary overvoltage inversion relationship is wrong, online experts intervene to construct a new overvoltage inversion relationship, and send the construction results and influencing factors to the machine learning module.

[0083] Among them, the influencing factors include the arrester status data, the arrester operation time, the number of lightning strikes the arrester has experienced, the arrester model, the number of arresters connected in series, the meteorological conditions in the area where the arrester is located, and the arrester initial status data.

[0084] In some embodiments, an inversion relationship creation system is also provided. After the manual construction module confirms the newly established inversion relationship, the inversion relationship is confirmed by the inversion relationship creation system, and the inversion relationship is fed back to the intelligent data analysis terminal and stored.

[0085] In some embodiments, the machine learning module is associated with the inversion relationship creation system, and both the inversion relationship constructed by the machine learning module itself and the inversion relationship constructed by the manual construction module are sent to the inversion relationship creation system to obtain a new inversion relationship, and the inversion relationship creation system and the inversion relationship creation module are communicatively connected, and the obtained new inversion relationship is sent to the inversion relationship creation module, and the new correct inversion relationship is stored in the inversion relationship storage module.

[0086] Among them, the intelligent data analysis terminal obtains a variety of data and also processes the collected data, establishes and stores the correspondence between the tail end current simulation data and the lightning simulation data, and stores the correspondence between the tail end current real data and the lightning real data.

[0087] Among them, the real lightning data and the real tail current data obtained by the current acquisition module are also obtained, and the data are preprocessed, including removing noise signals, converting the real tail current data into digital signals, and converting the lightning simulation data into digital signals.

[0088] It also includes a lightning acquisition module to determine whether lightning occurs in the current area when the lightning arrester is in operation. If so, the real lightning data is obtained at the same time.

[0089] Among them, because the signals obtained in different scenarios are different, when the arrester device is newly installed, the signal set at the pressure end is lightning simulation data. This signal is simulated by simulating the frequency, voltage and other data of lightning, and applied to the pressure end. When the tail current is generated, the tail current simulation data is obtained. After the arrester is installed in the power grid system, the pressure end is no longer in operation, and this part becomes the head end of the arrester, and the lightning acquisition module obtains the real lightning data.

[0090] Among them, after the tail current simulation data acquisition module and the lightning simulation data acquisition module obtain all the data, the data is sent to the inversion relationship creation module. When the lightning arrester has not yet been installed in the power system, the key data are the lightning simulation data and the tail current simulation data. A corresponding relationship is established between these two types of data, thus forming an inversion relationship between the arrester tail current simulation data and the lightning simulation data.

[0091] Among them, the intelligent data analysis terminal also obtains the data collected by the lightning collection module. After the lightning arrester is installed in the power system, when lightning occurs, the data of the lightning collection module is also used as the original data that needs to be set.

[0092] Among them, when the lightning arrester has not yet been installed in the power system, the data obtained from the pressure-applying end and the current acquisition module are used to establish the inverse relationship between the tail-end current frequency, the tail-end current and the head-end voltage, and to establish a database.

[0093] Among them, an inversion relationship storage module can be formed based on the inversion relationship creation module, but all the inversion relationships established in the data are realized based on lightning simulation data, and for some lightning data, it is impossible to simulate, that is, it is impossible to realize the simulation processing of all lightning signals. Obviously, in the inversion relationship, it will appear in the actual operation of the power system. If the collected real data of the tail current does not exist in the inversion relationship library, it will cause the inversion system to fail to operate normally. For this reason, an inversion relationship verification module is set.

[0094] Among them, after the inversion relationship verification module obtains the real data of the tail current transmitted by the tail current acquisition module, it determines from the inversion relationship that has been created whether the currently acquired tail current contains the corresponding overvoltage inversion data. If the corresponding overvoltage inversion data can be found therein, it can be considered that the overvoltage inversion relationship corresponding to the current already exists. If the current data does not have the corresponding overvoltage inversion data, it is necessary to construct the corresponding overvoltage inversion relationship for the real data of the tail current.

[0095] In some embodiments, in order to ensure the accuracy of the inversion relationship between the tail current and overvoltage of the installed arrester, the current arrester operation time, the number of lightning strikes sustained by the arrester during operation, the state of the arrester itself, etc. are also collected, and all these data are collected. At the same time, the lightning data obtained by the lightning acquisition module is obtained, and the relationship between the tail current and the lightning data is constructed, thereby forming a new inversion relationship.

[0096] Among them, after confirming that there is no corresponding inversion relationship in the database, the inversion relationship verification module will obviously establish a new inversion relationship. At this time, the obtained new inversion relationship will be sent to the machine learning module. The role of the machine learning module is to obtain the established new inversion relationship, and to perform machine learning based on the inversion database obtained by the inversion relationship creation module that has been constructed, to determine the relationship between lightning data and tail current of the same model of lightning arrester in the same operating time, and to perform preliminary verification of the inversion relationship. For example, in the data processing, the obtained data is subjected to a fast Fourier transform to obtain a simple corresponding relationship, and the machine learning module determines the overvoltage inversion result corresponding to the tail current currently collected based on the interpolation method for the tail current and overvoltage inversion results of the lightning arrester of the same model and the same operating time. At the same time, the overvoltage inversion result obtained by the machine learning module is compared with the lightning data collected by the lightning acquisition module. If the deviation of the two data is found to be too large, the obtained data will be sent to the manual construction module.

[0097] Among them, the role of the manual construction module is that after the machine learning module directly establishes the correspondence between the real lightning data and the real tail current data, the manual construction module intervenes to check and modify the problems existing in the current overvoltage inversion data processing process, and completes the task according to the experience of the manual construction module. For example, confirm whether the lightning arrester is damaged, whether there is a problem with the lightning arrester wiring, etc., and consider other parameters at the same time to jointly analyze the problems existing in the inversion relationship established by the current machine learning module. After the analysis, the new restrictions introduced by the manual construction module, the corrections to the algorithm, etc., are all sent back to the machine learning module for machine learning.

[0098] Among them, after the artificial construction module reconstructs the new tail current and overvoltage inversion relationship, the new tail current and overvoltage inversion relationship, the corrected algorithm, and the detection results of the lightning arrester are fed back to the machine learning module, so that in the subsequent work, if the machine learning module obtains the tail current data and overvoltage inversion relationship that cannot be verified by the inversion relationship, the machine learning module can gradually imitate the artificial construction module to construct a new tail current and overvoltage inversion relationship.

[0099] Among them, after the machine learning module collects the conclusions sent to it by the manual construction module, or the newly established inversion relationship is correct, the machine learning module sends the relevant results to the inversion relationship new creation system, and sends the obtained inversion results to the inversion relationship creation module, thereby adding a new inversion relationship to the inversion relationship storage module.

[0100] Embodiment 3

[0101] Figure 3 An operation flow chart of an overvoltage inversion system based on arrester tail current monitoring is provided in an embodiment of the present invention. This embodiment is applicable to the overvoltage inversion process of arrester tail current monitoring, wherein an overvoltage inversion relationship based on arrester tail current data is first established based on lightning simulation data and arrester tail current simulation data, and then a new inversion relationship is constructed based on a machine learning module and a manual construction module to supplement the inversion relationship. At the same time, the machine learning module can gradually establish the inversion relationship according to the inversion relationship establishment method of the manual construction module, so that the machine learning module can independently construct the correct inversion relationship based on the obtained data information.

[0102] Among them, it is necessary to configure a lightning arrester. In the current power grid, usually a lightning arrester cannot be set up to operate normally with a single unit, but multiple lightning arresters need to be connected in series to meet the normal operation requirements. Therefore, when using it, it is necessary to set the lightning arrester according to the lightning arrester setting method in the power grid system. For example, in the use of a certain type of lightning arrester, three lightning arresters are connected in series. In the construction of the tail end current and overvoltage inversion relationship of the lightning arrester, a pressure-enhancing end is set at the head end of the three series-connected lightning arresters, and the pressure-enhancing end applies lightning simulation data, that is, according to the voltage, frequency and other setting data of the lightning, the pressure setting of the lightning arrester is realized. At the same time, a current acquisition module is connected to the tail end of the lightning arrester, and the current acquisition module receives the current signal generated after the lightning simulation data is applied.

[0103] Among them, after the lightning arrester is configured, lightning simulation data is applied to the lightning arrester, and the current simulation data of the tail end of the lightning arrester is obtained at the same time. A corresponding relationship is established between the current simulation data of the tail end of the lightning arrester and the lightning simulation data, thus forming an inversion relationship.

[0104] In some embodiments, regarding the configuration of the lightning arrester, instead of simply constructing an inversion relationship for multiple lightning arresters in a serial connection state, an inversion relationship is also constructed for a single lightning arrester.

[0105] Among them, after the arrester is configured and the lightning simulation data is input, usually this type of inversion relationship is only for the arrester that has not been installed in the power grid system, or the arrester under normal operating conditions, and the arrester state after the arrester has been running for a period of time cannot be identified. At this time, for the established inversion relationship, it is also necessary to collect the inversion relationship of the arrester that has been officially applied for a period of time, and all inversion relationships need to be stored in the inversion relationship storage module.

[0106] In some embodiments, for lightning arresters that have been formally used for a period of time, an on-site analysis is performed on such lightning arresters, and an inversion relationship between the lightning arrester tail end current simulation data and the overvoltage is constructed by applying actual lightning simulation data.

[0107] In some embodiments, the arrester is processed by corrosion testing to fully simulate the real situation, and then lightning simulation data is applied to it to construct an inversion relationship between the arrester tail current simulation data and the overvoltage.

[0108] Among them, in the established inversion relationship storage module, the inversion relationship between the arrester tail current simulation data and the overvoltage under all conditions of lightning simulation data application is obtained.

[0109] Among them, during the official activation of the inversion system, the current acquisition module set on the lightning arrester is used to obtain the real data of the tail current generated when the lightning arrester is installed in the power system and in the lightning state. At the same time, the lightning acquisition module is used to obtain the lightning data generated at the top of the lightning arrester in the lightning state.

[0110] Among them, the collected actual tail current and the real lightning data generated at the top of the lightning arrester are pre-processed, including noise elimination, analog-to-digital conversion, etc.

[0111] Among them, after the tail current and lightning data are preprocessed, it is analyzed whether the corresponding overvoltage can be found in the established inversion relationship storage module for the current real data of the tail current. If it can be found, the overvoltage inversion result can be given directly according to the current inversion relationship. If the corresponding result cannot be found, it is considered that the current real data of the tail current is not included in the inversion relationship library, and it is necessary to construct a new inversion relationship based on this real value of the tail current, and update the inversion relationship in the inversion relationship storage module.

[0112] Among them, in the matching operation of the real lightning data and the inversion relationship, when the corresponding matching result cannot be retrieved from the database, the obtained real data of the tail current and the lightning data collected by the lightning collection module are sent to the machine learning module.

[0113] Among them, the machine learning module establishes new tail current real data and overvoltage inversion relationship based on the relevant algorithm of lightning data and inversion relationship matching relationship. The specific steps are as follows:

[0114] Step 1: Establish the initial inversion relationship.

[0115] During the operation of the arrester, the collected tail current is directly corresponding to the data obtained based on the lightning collection module, and the initial inversion relationship of the tail current is obtained. This process is directly constructed by the machine learning module and then verified by the machine learning module.

[0116] Step 2: Verify the initial inversion relationship.

[0117] For the initial inversion relationship that has been obtained, the machine learning module uses the interpolation method to obtain the data of the previous node and the next node respectively for the tail current and lightning data, and uses the interpolation method to determine the real data of the tail current of the current node and the lightning data corresponding to the current node obtained based on the interpolation method. Compare the lightning data corresponding to the current node obtained based on the interpolation method with the lightning data collected by the lightning acquisition module, and the comparison equation is as follows:

[0118]

[0119] Among them, σ represents the deviation value, U f Indicates the lightning data collected by the lightning collection module, U t It represents the lightning data corresponding to the current node obtained based on the interpolation method. The deviation value threshold is set to the range of (0, 5%]. When the calculation result is within this range, the lightning data obtained based on the interpolation method can be applied normally, and the established initial inversion relationship can be directly applied. If the calculation result exceeds this range, it means that the initial inversion relationship cannot be used directly and needs to be corrected.

[0120] Step 3: Manual construction module verification

[0121] When the machine learning module establishes the initial inversion relationship, and after verification, it is found that the initial inversion relationship is actually incorrect, because the machine learning module lacks the corresponding algorithm or does not consider the constraints sufficiently, in order to ensure that the newly established tail current real data and overvoltage inversion relationship are correct, the relevant data are submitted to the manual construction module. The manual construction module considers the application time of the arrester, the contamination of the arrester, the insulation of various types of arrester connections, etc., to establish a new arrester tail current real data and overvoltage inversion relationship, and the work results of the manual construction module are fed back to the machine learning module.

[0122] Step 4: Machine Learning Module Learning

[0123] When the machine learning module collects the feedback results given by the manual construction module, it will perform machine learning based on the inversion relationship construction results provided by the manual construction module and the collection method of relevant information, so that the machine learning module can gradually create a new inversion relationship between the real data of the arrester tail current and the overvoltage according to the working method of the manual construction module. For example: the manual construction module first analyzes whether the arrester is damaged, and the damage of the arrester is based on the large data fluctuations in the historical data of the arrester tail current. When it is found that there is a large data fluctuation, it directly gives the conclusion that the arrester is not applicable. When the subsequent machine learning module learns the working method of the manual construction module, after discovering the newly created inversion relationship between the real data of the arrester tail current and the overvoltage, it will give priority to obtaining the historical data of the real data of the arrester tail current, and analyze whether there is a similar large data fluctuation. If it is found that there is, the machine learning module directly issues an arrester damage alarm instead of trying to build a new inversion relationship between the real data of the arrester tail current and the overvoltage.

[0124] Among them, after constructing the new real data of the tail current of the lightning arrester and the overvoltage inversion relationship, the new inversion relationship is input into the inversion relationship storage module to supplement the inversion relationship in the database, and it can be created by itself based on the machine learning module.

[0125] Among them, for the work involving the machine learning module, the lightning arrester is actually already in actual application and operation. Therefore, when establishing a new inversion relationship between the tail current and overvoltage of the lightning arrester, it is also necessary to synchronously record the operating time of the lightning arrester.

[0126] In some embodiments, in addition to obtaining the operating time of the lightning arrester, the number of lightning occurrences of the lightning arrester during the operating cycle is also obtained synchronously. Therefore, when the machine learning module and the manual construction module construct a new tail current and overvoltage inversion relationship, data such as the number of lightning occurrences and the operating time of the lightning arrester can be introduced to obtain a more accurate and appropriate algorithm.

[0127] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. An overvoltage inversion system based on arrester tail current monitoring, characterized in that: include: A pressurizing end, mounted on the top of the arrester, for applying lightning simulation data to the arrester; A lightning collection module is installed on the top of the lightning arrester and is used to collect real lightning data; A current acquisition module, installed at the tail end of the arrester, for collecting the tail end current simulation data of the arrester and the tail end current real data of the arrester; The tail end current simulation data is the current data generated under the action of the lightning simulation data, and the tail end current real data is the current data generated under the action of the lightning real data; An intelligent data analysis terminal, connected to the current acquisition module, the lightning acquisition module and the pressurizing terminal respectively, for establishing and storing a correspondence between the tail end current simulation data and the lightning simulation data and storing a correspondence between the tail end current real data and the lightning real data; An inversion relationship verification module, connected to the intelligent data analysis terminal, for performing inversion and verification using the corresponding relationship between the tail end current simulation data and the lightning simulation data; The inversion relationship creation module is connected to the intelligent data analysis terminal and is used to establish the corresponding relationship between the real lightning data and the real tail current data after inversion and verification.

2. The overvoltage inversion system based on arrester tail current monitoring according to claim 1 is characterized in that: The intelligent data analysis terminal includes: A lightning simulation data acquisition module, connected to the pressurizing end, for acquiring the lightning simulation data; A tail end current simulation data acquisition module, connected to the current acquisition module, for acquiring the tail end current simulation data; An inversion relationship creation module, connected to the lightning simulation data acquisition module and the tail current simulation data acquisition module respectively, for establishing a corresponding relationship between the lightning simulation data and the tail current simulation data; The inversion relationship storage module is connected to the inversion relationship creation module and the inversion relationship verification module, and is used to store the corresponding relationship between the tail end current simulation data and the lightning simulation data and the corresponding relationship between the tail end current real data and the lightning real data.

3. The overvoltage inversion system based on arrester tail current monitoring according to claim 1 is characterized in that: The inversion relationship verification module includes: A tail end current real data acquisition unit, connected to the current acquisition module, for acquiring the tail end current real data; A lightning real data acquisition unit, connected to the lightning acquisition module, for acquiring the lightning real data; a first judgment unit connected to the tail current real data acquisition unit, and used to judge whether the tail current real data exists in the intelligent data analysis terminal; if it exists in the intelligent data analysis terminal, the lightning simulation data is inverted according to the corresponding relationship between the lightning simulation data and the tail current simulation data to obtain the lightning simulation inversion data; a deviation value calculation unit, connected to the lightning real data acquisition unit and the first judgment unit respectively, and used to calculate the deviation value between the lightning simulation inversion data and the lightning real data; The second judgment unit is connected to the deviation value calculation unit, and is used to judge whether the deviation value is within the deviation value threshold range. If it is within the deviation value threshold range, it means that the correspondence between the real data of the tail end current and the lightning simulation data has passed the verification, and the verification result is sent to the intelligent data analysis end for storage; if it is not within the deviation value threshold range, it means that the correspondence between the real data of the tail end current and the lightning simulation data has not passed the verification.

4. The overvoltage inversion system based on arrester tail current monitoring according to claim 3 is characterized in that: The system further comprises: A machine learning module is connected to the inversion relationship verification module, and is used to establish a corresponding relationship between the real data of the tail current and the real data of lightning when the corresponding relationship between the real data of the tail current and the lightning simulation data fails to pass the verification, so as to form a temporary overvoltage inversion relationship; An evaluation module, connected to the machine learning module, for determining whether the temporary overvoltage inversion relationship is correct; if the temporary overvoltage inversion relationship is correct, the temporary overvoltage inversion relationship is authenticated as a new overvoltage inversion relationship; if the temporary overvoltage inversion relationship is wrong, the temporary overvoltage inversion relationship is deleted; The manual construction module is connected to the machine learning module and is used for, when the temporary overvoltage inversion relationship is wrong, for online experts to intervene, to construct a new overvoltage inversion relationship, and to send the construction results and influencing factors to the machine learning module.

5. The overvoltage inversion system based on arrester tail current monitoring according to claim 4 is characterized in that: The influencing factors include arrester status data, arrester operation time, number of lightning strikes sustained by the arrester, arrester model, number of arresters connected in series, meteorological conditions in the area where the arrester is located, and arrester initial status data.

6. An overvoltage inversion method based on current monitoring at the tail end of a lightning arrester, used to implement an overvoltage inversion system based on current monitoring at the tail end of a lightning arrester as claimed in any one of claims 1 to 5, characterized in that: The steps include: Apply lightning simulation data to the arrester using the pressurized end; Use the lightning collection module installed on the top of the lightning arrester to collect real lightning data; Using a current collection module installed at the tail end of the arrester to collect the tail end current simulation data of the arrester and the tail end current real data of the arrester; The tail end current simulation data is the current data generated under the action of the lightning simulation data, and the tail end current real data is the current data generated under the action of the lightning real data; Using an intelligent data analysis terminal to establish and store a correspondence between the tail end current simulation data and the lightning simulation data, and to store a correspondence between the tail end current real data and the lightning real data; Using an inversion relationship verification module to perform inversion and verification using the corresponding relationship between the tail current simulation data and the lightning simulation data; The corresponding relationship between the real lightning data and the real tail current data is established by using the inversion relationship new creation module after the inversion and verification are passed.

7. The overvoltage inversion method based on arrester tail current monitoring according to claim 6 is characterized in that: The inversion relationship verification module uses the corresponding relationship between the tail current simulation data and the lightning simulation data to perform inversion and verification, specifically including: Determine whether the tail end current real data exists in the intelligent data analysis terminal; if it exists in the intelligent data analysis terminal, invert the lightning simulation data according to the corresponding relationship between the lightning simulation data and the tail end current simulation data to obtain lightning simulation inversion data; Calculating the deviation between the lightning simulation inversion data and the lightning real data; Determine whether the deviation value is within the deviation value threshold range. If so, it means that the correspondence between the real data of the tail end current and the lightning simulation data has been verified, and the verification result is sent to the intelligent data analysis end for storage; if not, it means that the correspondence between the real data of the tail end current and the lightning simulation data has not been verified.

8. An antenna interface unit test terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 6 to 7 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 6 to 7 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 6 to 7 is implemented.

Images