Lightning arrester on-line monitoring system and method based on simulation model

Abstract

The invention discloses a lightning arrester on-line monitoring system and method based on a simulation model, and belongs to the technical field of lightning arrester monitoring. Historical leakage current information and historical event triggering node time information of each historical event triggering node of the lightning arrester are called through log records of the lightning arrester; marking a historical event triggering state; based on the marking result of the event triggering state, performing mapping of historical event triggering node time information on the event triggering state; calculating an event triggering probability according to the mapping relation; constructing a three-dimensional model of the lightning arrester through a simulation technology, and calculating the similarity between the actual appearance characteristics and the standard appearance characteristics of the lightning arrester; predicting an abnormal risk value of the lightning arrester based on the current operation time of the lightning arrester; when the abnormal risk value of the lightning arrester is greater than or equal to the risk threshold value, early warning prompt information is sent to a worker; therefore, immediate response to emergencies is realized, and meanwhile, the stability of an important power system is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of arrester monitoring, and in particular to an arrester online monitoring system and method based on a simulation model. Background Art

[0002] Simulation model is an important research and testing tool. It can help researchers test and analyze the system by simulating and analyzing real objects or systems through physical models or digital models, saving time and cost. Physical models usually test and verify designs, simulate system behavior and predict performance through physical models. Mathematical models usually simulate system behavior, analyze the relationship between variables and make predictions through mathematical equations, formulas and charts. Researchers choose appropriate models based on the specific research objects and research purposes.

[0003] The monitoring frequency of lightning arresters is usually not fixed, but is determined by multiple factors such as the type of lightning arrester, the installation environment, and local climatic conditions. Common monitoring frequencies for lightning arresters are regular annual inspections, semi-annual inspections, quarterly inspections, and on-demand inspections, which cannot respond immediately to emergencies and may affect the safety and stability of important power systems. Summary of the invention

[0004] The object of the present invention is to provide a lightning arrester online monitoring system and method based on a simulation model to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A method for online monitoring of a lightning arrester based on a simulation model, the method comprising the following steps:

[0007] Step S1. Extract the log records of the arrester, retrieve the historical leakage current information and the historical event triggering node time information of each historical event triggering node of the arrester; mark the historical event triggering status according to the historical leakage current information and the historical event triggering node time information;

[0008] Step S2. Based on the marking result of the event trigger state, the event trigger state is mapped to the time information of the historical event trigger node; according to the mapping relationship, the event trigger probability is calculated;

[0009] Step S3. Construct a three-dimensional model of the arrester by simulation technology; based on the arrester simulation model, calculate the similarity between the actual appearance features of the arrester and the standard appearance features;

[0010] Step S4. Based on the current operating time of the lightning arrester, predict the abnormal risk value of the lightning arrester; when the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, send a warning prompt message to the staff.

[0011] Furthermore, the specific implementation process of step S1 includes:

[0012] Step S11. Collect the value of the arrester leakage current through the leakage current meter; record the arrester operation time through the time recorder, and generate a log record of the arrester;

[0013] Step S12. Mark the trigger status of historical events, obtain the time information of any historical event trigger node and the historical leakage current information corresponding to the time information of the historical event trigger node, the event trigger node is the node of the arrester trigger event status in the preset time period, and record an event trigger node at the end of each preset time period; mark the event trigger status of the historical event trigger node time information, and the event trigger status includes normal arrester, abnormal arrester and damaged arrester; in the historical leakage current information corresponding to the time information of the historical event trigger node, if the leakage current is less than the first preset leakage current, mark the time information of the historical event trigger node as abnormal arrester; if the leakage current is greater than the second preset leakage current, mark the time information of the historical event trigger node as damaged arrester; otherwise, mark the time information of the historical event trigger node as normal arrester; sort out and mark the event trigger status of all historical event trigger nodes.

[0014] As time goes by, the lightning arrester will age to a certain extent. The higher the degree of aging of the lightning arrester, the greater the leakage current. When the leakage current reaches the first preset leakage current, the arrester will function abnormally, requiring staff to check for risks. When the leakage current reaches the second preset leakage current, the arrester cannot protect electrical equipment from lightning overvoltage, operating overvoltage and power frequency transient overvoltage, requiring staff to replace the equipment.

[0015] Furthermore, the specific implementation process of step S2 includes:

[0016] Step S21. Count all the time information of the historical event triggering nodes and generate a set of the time of the historical event triggering nodes, denoted as T = {T m |m∈[1,M]}, where T m represents the time of the mth historical event triggering node, M represents the total number of historical event triggering node times; the event triggering state is marked as C = {C v |v∈[1,3]}, where C1, C2, and C3 represent normal arrester, abnormal arrester, and damaged arrester, respectively; the event trigger state is mapped to the historical event trigger node time → event trigger state, where one historical event trigger node time corresponds to one historical event trigger state, i.e., T m →Cv ;

[0017] Step S22: sort out the event triggering transfer situation. If the event triggering transfer situation occurs between two adjacent historical event triggering node times, mark the event as a triggering transfer event. The triggering transfer event is represented by C i →C j , where i, j∈[1,3] and i≠j, and generates an event-triggered transfer matrix, denoted as: {F ij} 3×3 ;

[0018] Step S23. Count the number of each type of triggering transfer event, recorded as SUPij, and calculate the triggering transfer probability: Among them, P ij Indicated by C i →C j The probability of triggering a transition event.

[0019] Due to the universality of lightning arresters, there are three possibilities for the arrester to transfer from a certain triggering state to another uncertain triggering state. The probability of each possibility in real life is different. i →C j Among three possible probabilities, it can be predicted that the next event triggers node C i →C j possibility, thereby predicting whether the lightning arrester is working properly.

[0020] Furthermore, the specific implementation process of step S3 includes:

[0021] Step S31. Based on the arrester log record, generate the arrester 3D model using simulation software; based on the arrester 3D model, extract the standard appearance features of the arrester 3D model in all directions, and construct a standard appearance feature set DES = {D a |a∈[1,A]}, where D a A pixel point set of the standard appearance features of the three-dimensional model in the a-th direction, where A represents the total number of directions of the three-dimensional model, and the pixel points include the coordinate points and pixel values ​​of the pixel points;

[0022] Step S32: extract the actual appearance features of the three-dimensional model in each direction of the arrester's historical event triggering node time, and construct an appearance feature set MES = {M a |a∈[1,A]}, where M a A set of pixels representing the actual appearance features of the 3D model in the ath direction;

[0023] Step S33. Calculate the similarity between the standard appearance feature and the actual appearance feature: Among them, NUM(Da ∩M a ) represents the number of pixels with the same pixel value at the same coordinate point in the three-dimensional model in the ath direction, NUM(D a ∪M a ) represents the total number of pixels of the three-dimensional model in the ath direction.

[0024] Whether the arrester needs to be checked requires not only an internal evaluation through the size of the leakage current, but also an external evaluation through whether the similarity between the actual appearance characteristics of the arrester and the standard appearance characteristics meets the standards. The standard appearance characteristics are the appearance characteristics of the arrester when it is not dirty or damaged. When the outside of the arrester is dirty, it will cause local heating on the surface of the arrester. This heating phenomenon not only accelerates the aging of the arrester, but may also affect the voltage distribution inside the arrester, causing some valve plates to be subjected to excessive voltage, further accelerating aging or causing flashover of insulating parts, thereby causing the leakage current to increase, affecting the judgment of the performance of the arrester.

[0025] Furthermore, the specific implementation process of step S4 includes:

[0026] Step S41. Based on the current operating time of the arrester, predict the probability of triggering transfer when the arrester is abnormal:

[0027] p=P 12 +P 13 ; Calculate the abnormal value d of the lightning arrester. When the trigger transfer probability of the abnormality of the lightning arrester reaches the preset abnormal threshold, the abnormal value d of the lightning arrester is 1, otherwise, the abnormal value d of the lightning arrester is 0;

[0028] When the trigger state changes from normal state to abnormal state and damaged state, it means that there is a certain problem with the lightning arrester. For example, the lightning arrester is dirty, resulting in increased leakage current. It only needs to be cleaned and the lightning arrester can work normally. If the leakage current is too large and the lightning arrester is damaged, the staff needs to directly replace the parts. These situations require timely investigation by the staff.

[0029] Step S42. Based on the current three-dimensional model of the arrester, calculate the similarity between the standard appearance features of the arrester and the actual appearance features;

[0030] Step S43. Calculate the abnormal risk value of the lightning arrester: f = d-sign (HUE-1), where sign (*) represents the sign function; preset a risk value threshold, and if the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, issue a warning message to the staff, the warning message includes the predicted trigger transfer probability of the lightning arrester abnormality and the similarity between the standard appearance features of the lightning arrester and the actual appearance features.

[0031] When the internal leakage current of the arrester may be abnormal but the appearance has not changed, that is, d=1, sign(HUE-1)=0, the staff is required to check the abnormality to see whether the internal leakage current is temporarily abnormal due to the replacement of parts or whether the internal function is damaged; when the internal leakage current of the arrester is normal but the appearance has changed (such as dirt and cracks), that is, d=0, sign(HUE-1)=-1, the staff is required to check the abnormality to see whether the arrester casing needs to be cleaned or repaired to prevent the casing from causing accelerated aging of the arrester; when the internal leakage current of the arrester is abnormal but the appearance has changed, that is, d=1, sign(HUE-1)=-1, the staff is required to check both the casing and the interior of the arrester.

[0032] A lightning arrester online monitoring system based on a simulation model, the system comprising a data acquisition module, an event triggering module, a feature similarity module and an early warning module;

[0033] The data acquisition module is used to retrieve the historical leakage current information and the historical event triggering node time information of each historical event triggering node of the arrester; and mark the historical event triggering state according to the historical leakage current information and the historical event triggering node time information;

[0034] The event trigger module maps the event trigger state to the time information of the historical event trigger node based on the marking result of the event trigger state; and calculates the event trigger probability according to the mapping relationship;

[0035] The feature similarity module constructs a three-dimensional model of the arrester through simulation technology; based on the arrester simulation model, calculates the similarity between the actual appearance features of the arrester and the standard appearance features;

[0036] The early warning module predicts the abnormal risk value of the lightning arrester based on the current operating time of the lightning arrester; when the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, a warning prompt message is issued to the staff.

[0037] Further, the data acquisition module includes a log recording unit and a marking unit;

[0038] The log recording unit collects the value of the arrester leakage current through the leakage current meter; records the arrester operation time through the time recorder, and generates a log record of the arrester;

[0039] The marking unit obtains the time information of any historical event triggering node and the historical leakage current information corresponding to the time information of the historical event triggering node by marking the triggering status of the historical event, wherein the event triggering node is a node of the arrester triggering event status in a preset time period, and an event triggering node is recorded at the end of each preset time period; the event triggering status is marked on the time information of the historical event triggering node, and the event triggering status includes normal arrester, abnormal arrester and damaged arrester; in the historical leakage current information corresponding to the time information of the historical event triggering node, if the leakage current is less than the first preset leakage current, the time information of the historical event triggering node is marked as abnormal arrester; if the leakage current is greater than the second preset leakage current, the time information of the historical event triggering node is marked as damaged arrester; otherwise, the time information of the historical event triggering node is marked as normal arrester; the event triggering status of all historical event triggering nodes is sorted out and marked.

[0040] Further, the event triggering module includes a mapping unit and a triggering transition probability unit;

[0041] The mapping unit collects all the historical event triggering node time information and generates a historical event triggering node time set, which is recorded as T={T m |m∈[1,M]}, where T m represents the time of the mth historical event triggering node, M represents the total number of historical event triggering node times; the event triggering state is marked as C = {C v |v∈[1,3]}, where C1, C2, and C3 represent normal arrester, abnormal arrester, and damaged arrester, respectively; the event trigger state is mapped to the historical event trigger node time → event trigger state, where one historical event trigger node time corresponds to one historical event trigger state, i.e., T m →C v ;

[0042] The triggering transfer probability unit sorts out the event triggering transfer situation. If an event triggering transfer situation occurs between two adjacent historical event triggering node times, the event is marked as a triggering transfer event. The triggering transfer event is represented by C i →C j , where i, j∈[1,3] and i≠j, and generates an event-triggered transfer matrix, denoted as: {F ij} 3×3 ; Count the number of each type of trigger transfer event, recorded as SUP ij , calculate the probability of triggering transition: Among them, P ij Indicated by C i →C jThe probability of triggering a transition event.

[0043] Further, the feature similarity module includes an appearance feature unit and a similarity unit;

[0044] The appearance feature unit generates a three-dimensional model of the arrester using simulation software based on the arrester log record; extracts standard appearance features of the arrester in various directions based on the arrester three-dimensional model, and constructs a standard appearance feature set DES = {D a |a∈[1,A]}, where D a The pixel point set of the standard appearance features of the three-dimensional model in the ath direction, A represents the total number of directions of the three-dimensional model, and the pixel point includes the coordinate point and pixel value of the pixel point; extract the actual appearance features of the three-dimensional model in each direction of the historical event triggering node time of the lightning arrester, and construct the appearance feature set MES={M a |a∈[1,A]}, where M a A set of pixels representing the actual appearance features of the 3D model in the ath direction;

[0045] The similarity unit is used to calculate the similarity between the standard appearance feature and the actual appearance feature: Among them, NUM(D a ∩M a ) represents the number of pixels with the same pixel value at the same coordinate point in the three-dimensional model in the ath direction, NUM(D a ∪M a ) represents the total number of pixels of the three-dimensional model in the ath direction.

[0046] Further, the early warning module includes a risk unit and an early warning unit;

[0047] The risk unit predicts the probability of triggering a transfer due to an abnormality of the arrester based on the current operating time of the arrester: p = P 12 +P 13 ; Calculate the abnormal value d of the lightning arrester. When the trigger transfer probability of the abnormality of the lightning arrester reaches the preset abnormal threshold, the abnormal value d of the lightning arrester is 1, otherwise, the abnormal value d of the lightning arrester is 0; Based on the current three-dimensional model of the lightning arrester, calculate the similarity between the standard appearance features of the lightning arrester and the actual appearance features; Calculate the abnormal risk value of the lightning arrester: f = d-sign(HUE-1), where sign(*) represents the sign function;

[0048] The early warning unit, by presetting a risk value threshold, sends an early warning message to the staff if the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold. The early warning message includes the predicted trigger transfer probability of the lightning arrester abnormality and the similarity between the standard appearance characteristics of the lightning arrester and the actual appearance characteristics.

[0049] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: in an online monitoring system and method for a lightning arrester based on a simulation model provided by the present invention, the historical leakage current information and the historical event triggering node time information of each historical event triggering node of the lightning arrester are retrieved through the log record of the lightning arrester; the historical event triggering state is marked; based on the marking result of the event triggering state, the event triggering state is mapped to the historical event triggering node time information; according to the mapping relationship, the event triggering probability is calculated; through the simulation technology, a three-dimensional model of the lightning arrester is constructed; based on the lightning arrester simulation model, the similarity between the actual appearance features of the lightning arrester and the standard appearance features is calculated; based on the current operating time of the lightning arrester, the abnormal risk value of the lightning arrester is predicted; when the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, an early warning prompt message is issued to the staff; thereby achieving an immediate response to emergencies and ensuring the stability of important power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0051] Figure 1 It is a structural schematic diagram of an arrester online monitoring system based on a simulation model of the present invention;

[0052] Figure 2 The present invention is a schematic diagram of the steps of an online monitoring method for a lightning arrester based on a simulation model. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] See also Figure 1-Figure 2 , the present invention provides a technical solution:

[0055] A method for online monitoring of a lightning arrester based on a simulation model, the method comprising the following steps:

[0056] Step S1. Extract the log records of the arrester, retrieve the historical leakage current information and the historical event triggering node time information of each historical event triggering node of the arrester; mark the historical event triggering status according to the historical leakage current information and the historical event triggering node time information;

[0057] Step S2. Based on the marking result of the event trigger state, the event trigger state is mapped to the time information of the historical event trigger node; according to the mapping relationship, the event trigger probability is calculated;

[0058] Step S3. Construct a three-dimensional model of the arrester by simulation technology; based on the arrester simulation model, calculate the similarity between the actual appearance features of the arrester and the standard appearance features;

[0059] Step S4. Based on the current operating time of the lightning arrester, predict the abnormal risk value of the lightning arrester; when the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, send a warning prompt message to the staff.

[0060] Specifically, the specific implementation process of step S1 includes:

[0061] Step S11. Collect the value of the arrester leakage current through the leakage current meter; record the arrester operation time through the time recorder, and generate a log record of the arrester;

[0062] Step S12. Mark the trigger status of historical events, obtain the time information of any historical event trigger node and the historical leakage current information corresponding to the time information of the historical event trigger node, the event trigger node is the node of the arrester trigger event status in the preset time period, and record an event trigger node at the end of each preset time period; mark the event trigger status of the historical event trigger node time information, and the event trigger status includes normal arrester, abnormal arrester and damaged arrester; in the historical leakage current information corresponding to the time information of the historical event trigger node, if the leakage current is less than the first preset leakage current, mark the time information of the historical event trigger node as abnormal arrester; if the leakage current is greater than the second preset leakage current, mark the time information of the historical event trigger node as damaged arrester; otherwise, mark the time information of the historical event trigger node as normal arrester; sort out and mark the event trigger status of all historical event trigger nodes.

[0063] Specifically, the specific implementation process of step S2 includes:

[0064] Step S21. Count all the time information of the historical event triggering nodes and generate a set of the time of the historical event triggering nodes, denoted as T = {T m |m∈[1,M]}, where T m represents the time of the mth historical event triggering node, M represents the total number of historical event triggering node times; the event triggering state is marked as C = {C v|v∈[1,3]}, where C1, C2, and C3 represent normal arrester, abnormal arrester, and damaged arrester, respectively; the event trigger state is mapped to the historical event trigger node time → event trigger state, where one historical event trigger node time corresponds to one historical event trigger state, i.e., T m →C v ;

[0065] Step S22: sort out the event triggering transfer situation. If the event triggering transfer situation occurs between two adjacent historical event triggering node times, mark the event as a triggering transfer event. The triggering transfer event is represented by C i →C j , where i, j∈[1,3] and i≠j, and generates an event-triggered transfer matrix, denoted as: {F ij} 3×3 ;

[0066] Step S23. Count the number of each type of trigger transfer event, recorded as SUP ij , calculate the probability of triggering transition: Among them, P ij Indicated by C i →C j The probability of triggering a transition event.

[0067] For example: the probability of a triggering transition event of a lightning arrester from a normal state to an abnormal state:

[0068] The probability of triggering the transfer event of the arrester from the normal state to the damaged state is:

[0069] The probability of a triggering transfer event of the arrester from normal state to normal state is:

[0070] Specifically, the specific implementation process of step S3 includes:

[0071] Step S31. Based on the arrester log record, generate the arrester 3D model using simulation software; based on the arrester 3D model, extract the standard appearance features of the arrester 3D model in all directions, and construct a standard appearance feature set DES = {D a |a∈[1,A]}, where D a The pixel point set of the standard appearance features of the 3D model in the ath direction, A represents the total number of directions of the 3D model, and the pixel points include the coordinate points and pixel values ​​of the pixel points;

[0072] Step S32: extract the actual appearance features of the three-dimensional model in each direction of the arrester's historical event triggering node time, and construct an appearance feature set MES = {M a|a∈[1,A]}, where M a A set of pixels representing the actual appearance features of the 3D model in the ath direction;

[0073] Step S33. Calculate the similarity between the standard appearance feature and the actual appearance feature: Among them, NUM(D a ∩M a ) represents the number of pixels with the same pixel value at the same coordinate point in the three-dimensional model in the ath direction, NUM(D a ∪M a ) represents the total number of pixels of the three-dimensional model in the ath direction.

[0074] Specifically, the specific implementation process of step S4 includes:

[0075] Step S41. Based on the current operating time of the arrester, predict the probability of triggering the arrester to fail: p = P 12 +P 13 ; Calculate the abnormal value d of the lightning arrester. When the trigger transfer probability of the abnormality of the lightning arrester reaches the preset abnormal threshold, the abnormal value d of the lightning arrester is 1, otherwise, the abnormal value d of the lightning arrester is 0;

[0076] Step S42. Based on the current three-dimensional model of the arrester, calculate the similarity between the standard appearance features of the arrester and the actual appearance features;

[0077] Step S43. Calculate the abnormal risk value of the lightning arrester: f = d-sign (HUE-1), where sign (*) represents the sign function; preset a risk value threshold, and if the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, issue a warning message to the staff, the warning message includes the predicted trigger transfer probability of the lightning arrester abnormality and the similarity between the standard appearance features of the lightning arrester and the actual appearance features.

[0078] An online monitoring system for lightning arresters based on a simulation model, the system comprises a data acquisition module, an event triggering module, a feature similarity module and an early warning module;

[0079] A data acquisition module is used to retrieve the historical leakage current information and the historical event triggering node time information of each historical event triggering node of the arrester; and mark the historical event triggering state according to the historical leakage current information and the historical event triggering node time information;

[0080] The event trigger module maps the event trigger state to the time information of the historical event trigger node based on the marking result of the event trigger state; and calculates the event trigger probability according to the mapping relationship;

[0081] The feature similarity module uses simulation technology to build a three-dimensional model of the arrester; based on the arrester simulation model, it calculates the similarity between the actual appearance features of the arrester and the standard appearance features;

[0082] The early warning module predicts the abnormal risk value of the lightning arrester based on the current operating time of the lightning arrester; when the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, an early warning message is issued to the staff.

[0083] Preferably, the data acquisition module includes a log recording unit and a marking unit;

[0084] The log recording unit collects the value of the arrester leakage current through the leakage current meter; records the arrester operation time through the time recorder, and generates a log record of the arrester;

[0085] A marking unit, by marking the trigger status of historical events, obtains the time information of any historical event trigger node and the historical leakage current information corresponding to the time information of the historical event trigger node, wherein the event trigger node is a node of the arrester trigger event status in a preset time period, and an event trigger node is recorded at the end of each preset time period; the event trigger status is marked for the time information of the historical event trigger node, and the event trigger status includes normal arrester, abnormal arrester and damaged arrester; in the historical leakage current information corresponding to the time information of the historical event trigger node, if the leakage current is less than the first preset leakage current, the time information of the historical event trigger node is marked as abnormal arrester; if the leakage current is greater than the second preset leakage current, the time information of the historical event trigger node is marked as damaged arrester; otherwise, the time information of the historical event trigger node is marked as normal arrester; the event trigger status of all historical event trigger nodes is sorted out and marked.

[0086] Preferably, the event trigger module includes a mapping unit and a trigger transition probability unit;

[0087] The mapping unit collects all the historical event trigger node time information and generates a historical event trigger node time set, denoted as T = {T m |m∈[1,M]}, where T m represents the time of the mth historical event triggering node, M represents the total number of historical event triggering node times; the event triggering state is marked as C = {C v |v∈[1,3]}, where C1, C2, and C3 represent normal arrester, abnormal arrester, and damaged arrester, respectively; the event trigger state is mapped to the historical event trigger node time → event trigger state, where one historical event trigger node time corresponds to one historical event trigger state, i.e., T m →C v ;

[0088] The trigger transfer probability unit sorts out the event trigger transfer situation. If an event trigger transfer situation occurs between two adjacent historical event trigger node times, the event is marked as a trigger transfer event. The trigger transfer event is represented by C i →C j , where i, j∈[1,3] and i≠j, and generates an event-triggered transfer matrix, denoted as: {F ij} 3×3 ; Count the number of each type of trigger transfer event, recorded as SUP ij , calculate the probability of triggering transition: Among them, P ij Indicated by C i →C j The probability of triggering a transition event.

[0089] Preferably, the feature similarity module includes an appearance feature unit and a similarity unit;

[0090] The appearance feature unit generates a three-dimensional model of the arrester based on the arrester log record using simulation software; based on the arrester three-dimensional model, extracts the standard appearance features of the arrester three-dimensional model in all directions, and constructs a standard appearance feature set DES = {D a |a∈[1,A]}, where D a The pixel point set of the standard appearance features of the three-dimensional model in the ath direction, A represents the total number of directions of the three-dimensional model, and the pixel point includes the coordinate point and pixel value of the pixel point; extract the actual appearance features of the three-dimensional model in each direction of the historical event triggering node time of the lightning arrester, and construct the appearance feature set MES = {M a |a∈[1, A]}, where Ma represents the pixel point set of the actual appearance features of the 3D model in the ath direction;

[0091] Similarity unit, used to calculate the similarity between the standard appearance features and the actual appearance features: Among them, NUM(D a ∩M a ) represents the number of pixels with the same pixel value at the same coordinate point in the three-dimensional model in the ath direction, NUM(D a ∪M a ) represents the total number of pixels of the three-dimensional model in the ath direction.

[0092] Preferably, the early warning module includes a risk unit and an early warning unit;

[0093] The risk unit predicts the probability of triggering an abnormality in the arrester based on the current operating time of the arrester: p = P 12 +P 13; Calculate the abnormal value d of the lightning arrester. When the trigger transfer probability of the abnormality of the lightning arrester reaches the preset abnormal threshold, the abnormal value d of the lightning arrester is 1, otherwise, the abnormal value d of the lightning arrester is 0; Based on the current three-dimensional model of the lightning arrester, calculate the similarity between the standard appearance features of the lightning arrester and the actual appearance features; Calculate the abnormal risk value of the lightning arrester: f = d-sign(HUE-1), where sign(*) represents the sign function;

[0094] The early warning unit presets a risk value threshold. If the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, an early warning message is issued to the staff. The early warning message includes the predicted trigger transfer probability of the lightning arrester abnormality and the similarity between the standard appearance characteristics of the lightning arrester and the actual appearance characteristics.

[0095] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0096] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A lightning arrester online monitoring method based on a simulation model, characterized in that: The method comprises the following steps: Step S1. Extract the log records of the arrester, retrieve the historical leakage current information and the historical event triggering node time information of each historical event triggering node of the arrester; mark the historical event triggering status according to the historical leakage current information and the historical event triggering node time information; Step S2. Based on the marking result of the event trigger state, the event trigger state is mapped to the time information of the historical event trigger node; according to the mapping relationship, the event trigger probability is calculated; Step S3. construct a three-dimensional model of the arrester by simulation technology; based on the arrester simulation model, calculate the similarity between the actual appearance features of the arrester and the standard appearance features; Step S4. Based on the current operating time of the lightning arrester, predict the abnormal risk value of the lightning arrester; when the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, send a warning prompt message to the staff.

2. The method for online monitoring of lightning arresters based on a simulation model according to claim 1, characterized in that: The specific implementation process of step S1 includes: Step S11. Collect the value of the arrester leakage current through the leakage current meter; record the arrester operation time through the time recorder, and generate a log record of the arrester; Step S12. Mark the trigger status of historical events, obtain the time information of any historical event trigger node and the historical leakage current information corresponding to the time information of the historical event trigger node, the event trigger node is the node of the arrester trigger event status in the preset time period, and record an event trigger node at the end of each preset time period; mark the event trigger status of the historical event trigger node time information, and the event trigger status includes normal arrester, abnormal arrester and damaged arrester; in the historical leakage current information corresponding to the time information of the historical event trigger node, if the leakage current is less than the first preset leakage current, mark the time information of the historical event trigger node as abnormal arrester; if the leakage current is greater than the second preset leakage current, mark the time information of the historical event trigger node as damaged arrester; otherwise, mark the time information of the historical event trigger node as normal arrester; sort out and mark the event trigger status of all historical event trigger nodes.

3. The method for online monitoring of lightning arresters based on a simulation model according to claim 2, characterized in that: The specific implementation process of step S2 includes: Step S21. Count all the time information of the historical event triggering nodes and generate a set of the time of the historical event triggering nodes, denoted as T = {T m |m∈[1,m]}, where T m represents the time of the mth historical event triggering node, M represents the total number of historical event triggering node times; the event triggering state is marked as C = {C v |v∈[1,3]}, where C1, C2, and C3 represent normal arrester, abnormal arrester, and damaged arrester, respectively; the event trigger state is mapped to the historical event trigger node time → event trigger state, where one historical event trigger node time corresponds to one historical event trigger state, i.e., T m →C v ; Step S22: sort out the event triggering transfer situation. If the event triggering transfer situation occurs between two adjacent historical event triggering node times, mark the event as a triggering transfer event. The triggering transfer event is represented by C i →C j , where i, j∈[1,3] and i≠j, and generates an event-triggered transfer matrix, denoted as: {F ij } 3×3 ; Step S23. Count the number of each type of trigger transfer event, recorded as SUP ij , calculate the probability of triggering transition: Among them, P ij Indicated by C i →C j The probability of triggering a transition event.

4. The method for online monitoring of lightning arresters based on a simulation model according to claim 3 is characterized in that: The specific implementation process of step S3 includes: Step S31. Based on the arrester log record, generate the arrester 3D model using simulation software; based on the arrester 3D model, extract the standard appearance features of the arrester 3D model in all directions, and construct a standard appearance feature set DES = {D a |a∈[1,A]}, where D a A pixel point set of the standard appearance features of the three-dimensional model in the a-th direction, where A represents the total number of directions of the three-dimensional model, and the pixel points include the coordinate points and pixel values ​​of the pixel points; Step S32: extract the actual appearance features of the three-dimensional model in each direction of the arrester's historical event triggering node time, and construct an appearance feature set MES = {M a |a∈[1,A]}, where M a A set of pixels representing the actual appearance features of the 3D model in the ath direction; Step S33. Calculate the similarity between the standard appearance feature and the actual appearance feature: Among them, NUM(D a ∩M a ) represents the number of pixels with the same pixel value at the same coordinate point in the three-dimensional model in the ath direction, NUM(D a ∪M a ) represents the total number of pixels of the three-dimensional model in the ath direction.

5. The method for online monitoring of lightning arresters based on a simulation model according to claim 4, characterized in that: The specific implementation process of step S4 includes: Step S41. Based on the current operating time of the arrester, predict the probability of triggering the arrester to fail: p = P 12 +P 13 ; Calculate the abnormal value d of the lightning arrester. When the trigger transfer probability of the abnormality of the lightning arrester reaches the preset abnormal threshold, the abnormal value d of the lightning arrester is 1, otherwise, the abnormal value d of the lightning arrester is 0; Step S42. Based on the current three-dimensional model of the arrester, calculate the similarity between the standard appearance features of the arrester and the actual appearance features; Step S43. Calculate the abnormal risk value of the lightning arrester: f = d-sign (HUE-1), where sign (*) represents the sign function; preset a risk value threshold, and if the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, issue a warning message to the staff, the warning message includes the predicted trigger transfer probability of the lightning arrester abnormality and the similarity between the standard appearance features of the lightning arrester and the actual appearance features.

6. A lightning arrester online monitoring system based on a simulation model, characterized in that: The system includes a data acquisition module, an event triggering module, a feature similarity module and an early warning module; The data acquisition module is used to retrieve the historical leakage current information and the historical event triggering node time information of each historical event triggering node of the arrester; and mark the historical event triggering state according to the historical leakage current information and the historical event triggering node time information; The event trigger module maps the event trigger state to the time information of the historical event trigger node based on the marking result of the event trigger state; According to the mapping relationship, the event triggering probability is calculated; The feature similarity module constructs a three-dimensional model of the arrester through simulation technology; Based on the arrester simulation model, the similarity between the actual appearance features of the arrester and the standard appearance features is calculated; The early warning module predicts the abnormal risk value of the lightning arrester based on the current operation time of the lightning arrester; When the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold, an early warning message is sent to the staff.

7. The arrester online monitoring system based on simulation model according to claim 6 is characterized in that: The data acquisition module includes a log recording unit and a marking unit; The log recording unit collects the value of the arrester leakage current through the leakage current meter; records the arrester operation time through the time recorder, and generates a log record of the arrester; The marking unit obtains the time information of any historical event triggering node and the historical leakage current information corresponding to the time information of the historical event triggering node by marking the historical event triggering state, wherein the event triggering node is the node of the arrester triggering event state in the preset time period, and an event triggering node is recorded at the end of each preset time period; Marking the event triggering status of the historical event triggering node time information, wherein the event triggering status includes normal lightning arrester, abnormal lightning arrester and damaged lightning arrester; In the historical leakage current information corresponding to the time information of the historical event triggering node, if the leakage current is less than the first preset leakage current, marking the time information of the historical event triggering node as an arrester abnormality; If the leakage current is greater than the second preset leakage current, the triggering node time information of the historical event is marked as a lightning arrester damage; Otherwise, the time information of the triggering node of the historical event is marked as the arrester is normal; The event triggering status of all historical event triggering nodes is sorted out and marked.

8. The arrester online monitoring system based on simulation model according to claim 7 is characterized in that: The event trigger module includes a mapping unit and a trigger transition probability unit; The mapping unit collects all the historical event triggering node time information and generates a historical event triggering node time set, which is recorded as T={T m |m∈[1,M]}, where T m represents the time of the mth historical event triggering node, M represents the total number of historical event triggering node times; the event triggering state is marked as C = {C v |v∈[1,3]}, where C1, C2, and C3 represent normal arrester, abnormal arrester, and damaged arrester, respectively; the event trigger state is mapped to the historical event trigger node time → event trigger state, where one historical event trigger node time corresponds to one historical event trigger state, i.e., T m →C v ; The triggering transfer probability unit sorts out the event triggering transfer situation. If an event triggering transfer situation occurs between two adjacent historical event triggering node times, the event is marked as a triggering transfer event. The triggering transfer event is represented by C i →C j , where i, j∈[1,3] and i≠j, and generates an event-triggered transfer matrix, denoted as: {F ij } 3×3 ; Count the number of each type of trigger transfer event, recorded as SUP ij , calculate the probability of triggering transition: Among them, P ij Indicated by C i →C j The probability of triggering a transition event.

9. The arrester online monitoring system based on simulation model according to claim 8, characterized in that: The feature similarity module includes an appearance feature unit and a similarity unit; The appearance feature unit generates a three-dimensional model of the arrester using simulation software based on the arrester log record; extracts standard appearance features of the arrester in various directions based on the arrester three-dimensional model, and constructs a standard appearance feature set DES = {D a |a∈[1,A]}, where D a The pixel point set of the standard appearance features of the three-dimensional model in the ath direction, A represents the total number of directions of the three-dimensional model, and the pixel point includes the coordinate point and pixel value of the pixel point; extract the actual appearance features of the three-dimensional model in each direction of the historical event triggering node time of the lightning arrester, and construct the appearance feature set MES={M a |a∈[1,A]}, where M a A set of pixels representing the actual appearance features of the 3D model in the ath direction; The similarity unit is used to calculate the similarity between the standard appearance feature and the actual appearance feature: Among them, NUM(D a ∪M a ) represents the number of pixels with the same pixel value at the same coordinate point in the three-dimensional model in the ath direction, NUM(D a ∪M a ) represents the total number of pixels of the three-dimensional model in the ath direction.

10. The arrester online monitoring system based on simulation model according to claim 9, characterized in that: The early warning module includes a risk unit and an early warning unit; The risk unit predicts the probability of triggering a transfer due to an abnormality of the arrester based on the current operating time of the arrester: p = P 12 +P 13 ; Calculate the abnormal value d of the lightning arrester. When the trigger transfer probability of the abnormality of the lightning arrester reaches the preset abnormal threshold, the abnormal value d of the lightning arrester is 1, otherwise, the abnormal value d of the lightning arrester is 0; Based on the current three-dimensional model of the lightning arrester, calculate the similarity between the standard appearance features of the lightning arrester and the actual appearance features; Calculate the abnormal risk value of the lightning arrester: f = d-sign (HUE-n), where sign (*) represents a sign function; The early warning unit, by presetting a risk value threshold, sends an early warning message to the staff if the abnormal risk value of the lightning arrester is greater than or equal to the risk value threshold. The early warning message includes the predicted trigger transfer probability of the lightning arrester abnormality and the similarity between the standard appearance characteristics of the lightning arrester and the actual appearance characteristics.

Images