A method and system for detecting the state of key components of an optical current transformer

Abstract

The application discloses a kind of optical current transformer key component state detection method and system, method includes: the CRS t Statistics and its statistical threshold value CRS c , carry out online self-detection;Establish the time series of the CRS t Statistics of representing the overall error state of three optical current transformers under the same node, calculate the fractal dimension of the time series, when the overall error state of three optical current transformers under the same node appears abnormal, according to the difference of fractal dimension, judge the degradation form of the overall error state of three optical current transformers under the same node;According to the degradation form, combine fault tree model, the abnormal influence source of optical current transformer measurement error is inversed, and the autonomous identification of optical current transformer fault position is realized;The application has the advantages that: method accurately deduces the abnormal influence source of optical current transformer measurement error, avoids security risks.

Description

Technical Field

[0001] This invention relates to the field of ultra-high voltage power technology, specifically to a method and system for detecting the status of key components of an optical current transformer. Background Technology

[0002] Optical current transformers, due to their advantages such as being oil-free and gas-free, safe and environmentally friendly, having a fast response speed, and a large dynamic range, have become the main development direction for current measurement and play an irreplaceable role in power systems, especially in ultra-high voltage direct current (UHVDC) transmission projects. However, in actual engineering operation, optical current transformers are susceptible to the coupling effects of various factors such as temperature, humidity, vibration, and electromagnetic interference. The operating performance of key components is prone to aging, which can lead to a decrease in the measurement accuracy of optical current transformers or even invalid output data, posing a significant threat to the safe and stable operation of UHVDC transmission systems. The traditional verification method based on periodic power outages and comparisons with standard current transformers, as stipulated in the verification procedure of "JJG 313 2010 Current Transformers for Measurement," suffers from problems such as difficulties in manufacturing standard current transformers, difficulties in outputting large currents on-site, and untimely error detection, resulting in insufficient reliability, flexibility, and economy in error detection for optical current transformers. On the other hand, traditional methods based on periodic comparisons of standard current transformers are a form of time-domain sampling inspection. The results of error detection are highly random and cannot further analyze the causes of error degradation in optical current transformers. For example, under the influence of random sources, the error of optical current transformers may exhibit random deviations, offering limited guidance for developing further maintenance and repair strategies. Effective identification of abnormal influence sources in optical current transformers is needed to guide more efficient and detailed maintenance and repair strategies.

[0003] Compared to other types of UHVDC equipment, optical current transformers are essentially signal sensing devices. Their key components are susceptible to performance degradation due to various factors, leading to decreased measurement errors. Furthermore, the forms of error degradation vary depending on the factors. Therefore, a method is urgently needed to analyze the forms of measurement error degradation caused by different factors in optical current transformers, thereby deduce the abnormal sources of measurement error. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that the existing methods for detecting the operating status of key components of optical current transformers cannot further analyze the causes of the error degradation of optical current transformers, thus failing to accurately deduce the abnormal influence sources of measurement errors of optical current transformers, which poses a safety hazard.

[0005] This invention solves the above-mentioned technical problems through the following technical means: a method for detecting the status of key components of an optical current transformer, comprising:

[0006] Step 1: Synchronously collect measurement data from three optical current transformers at the same node of the converter station, and calculate the CRS, which characterizes the overall error state of the three optical current transformers at the same node. t Statistical measure and its statistical threshold CRS c Online self-detection of the overall error status of three optical current transformers at the same node;

[0007] Step 2: Establish a CRS characterizing the overall error state of the three optical current transformers at the same node. t The time series of statistics is calculated, and the fractal dimension of the time series is determined. When the overall error state of three optical current transformers at the same node is abnormal, the deterioration mode of the overall error state of the three optical current transformers at the same node is determined according to the different fractal dimensions.

[0008] Step 3: Based on the overall error state degradation of the three optical current transformers at the same node, and combined with the fault tree model of the measurement error degradation characteristics of the optical current transformers, the abnormal influence sources of the optical current transformer measurement error are deduced.

[0009] Further, step one includes:

[0010] S11. Collect measurement data from three optical current transformers at the same node of the converter station. t =(I 1t ,I 2t ,I 3t ), where t is the sampling time of the optical current transformer measurement data, I 1t ,I 2t ,I 3t These are the current amplitude data corresponding to three optical current transformers at the same node;

[0011] S12. Measurement data x based on three optical current transformers at the same node t Calculate the CRS statistic, which characterizes the overall error state of the three optical current transformers at the same node at the sampling time, and the statistical threshold CRS used to measure whether the error exceeds the tolerance. c ;

[0012] S13. When sampling time t > n, if CRS t ≤CRS c The optical current transformer is in normal operating condition; if CRS t >CRS c, the optical current transformer operates abnormally under the interference of abnormal influencing sources, indicating that there is an optical current transformer with abnormal key components among the three optical current transformers. By calculating the contribution rate of the measurement data of the three optical current transformers to the statistical characteristic quantity SPE, the specific abnormal optical current transformer is judged, and the optical current transformer with the largest contribution rate is considered as the abnormal optical current transformer.

[0013] Furthermore, the S12 includes:

[0014] S121. Select the measurement data in the time period of 0 < t ≤ n to form a data matrix Perform singular value decomposition on the covariance matrix R of the data matrix X to obtain: R = X T X / (n - 1) = [P1 P2 P3]Λ[P1 P2 P3] T

[0015] In the formula, Λ = diag(λ1, λ2, λ3), λ1 ≥ λ2 ≥ λ3, the vectors P1, P2, P3 are the eigenvectors corresponding to the eigenvalues λ1, λ2, λ3; Λ represents the diagonal matrix about the eigenvalues λ1, λ2, λ3, diag represents the diagonal matrix, and λ1, λ2, λ3 represent the eigenvalues of the matrix R;

[0016] S122. Calculate the Hotelling T 2 statistic and SPE statistic of the data matrix X respectively, and the calculation methods are as follows:

[0017] In the formula, P e = [P2 P3];

[0018] S123. Calculate the expected values E(T 2 ), E(SPE) of the statistical characteristic quantities Hotelling T 2 and SPE respectively, and their overall standard deviations σ(T 2 ), σ(SPE);

[0019] S124. Calculate the CRS t statistic representing the overall error state of the three optical current transformers at the same node at the sampling time t (t > 0), and the calculation method is:

[0020] S125. Select the CRS t statistic representing the overall error state of the three optical current transformers at the same node in the time period of 0 < t ≤ n, and calculate the expected value E(CRS) and the overall standard deviation σ(CRS) of the statistical characteristic quantity CRS t ;

[0021] S126. Calculate the statistical threshold CRS that characterizes whether the overall error state of the three optical current transformers under the same node exceeds the tolerance based on the expected value E(Q) and the overall standard deviation σ(Q). c = E(CRS) + 3σ(CRS).

[0022] Furthermore, the calculation process of the contribution rate of the measurement data of the three optical current transformers to the statistical feature quantity SPE in S13 is as follows:

[0023]

[0024] In the formula, J it is the contribution rate of the measurement data of the i-th optical current transformer at the sampling time t to the modulus value SPE t ; x it represents the measurement data of the abnormal optical current transformer at time t; represents the projection of the measurement data of the abnormal optical current transformer at time t in the principal component subspace, where P represents the eigenvector P1 corresponding to the largest eigenvalue λ1 of the matrix R.

[0025] Further, the second step includes:

[0026] S21. Select the first n sampling times, establish the time series {CRS1, CRS2,... CRS n} of the CRS statistic, and calculate the fractal dimension FD0 of this time series;

[0027] S22. When the sampling time t > n, and CRS t > CRS c , establish the time series {CRS t , CRS t+1 ,... CRS t+9} of the CRS statistic, and calculate the fractal dimension FD t ;

[0028] S23. When |FD t - FD0| / FD0 * 100% > 1%, and FD t < FD0, determine that the measurement error of the abnormal optical current transformer is the ratio deviation A1 or the drift deviation A2. At this time, calculate the time series {CRS1, CRS2,... CRS n}, {CRS t , CRS t+1 ,... CRS t+9The standard deviations of the abnormal optical current transformer are S0 and S1; when |S1-S0| / S0*100%≤5%, the measurement error of the abnormal optical current transformer is determined to be the ratio deviation A1; when |S1-S0| / S0*100%>5%, the measurement error of the abnormal optical current transformer is determined to be the drift deviation A2.

[0029] When |FD t -FD0| / FD s *100%>1%, and FD t When >FD0, the measurement error of the abnormal optical current transformer is determined to be a random error increase of A3;

[0030] When |FD t -FD0| / FD0*100%≤1%, the measurement error of the abnormal optical current transformer is determined to be a fixed deviation A4.

[0031] Furthermore, step three includes:

[0032] S31, When sampling time t > n and CRS t >CRS c At that time, the time series of its fractal dimension {FD} is established. t ,FD t+1 ,...,FD t+9}, perform stationarity tests on data sequences with fractal dimensions;

[0033] If the data sequence of fractal dimension is stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a constant influencing factor T1, and the fault location of the abnormal optical current transformer needs to be further located.

[0034] If the data sequence of fractal dimension is non-stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a non-constant influencing factor T2, and there is no need to further locate the fault location of the abnormal optical current transformer.

[0035] S32. When the influencing factor of the measurement error of the abnormal optical current transformer is a constant influencing factor T1, it is necessary to locate the fault location of the abnormal optical current transformer based on the classification of fractal dimension and the fault tree model of the optical current transformer.

[0036] Furthermore, S31 includes:

[0037] S311, Calculate the time series {FD t ,FD t+1 ,...,FD t+9 The mean of} Through formula Construct the statistic Z, where E = 2N1N2 / 10 + 1, D = 2N1N2(2N1N2-10) / 900, and N1 is the fractal dimension of the time series less than 1. The number of elements, N² is the fractal dimension of the time series greater than 1. The number of;

[0038] S322. If |Z| < 1.96, the time series {FD} t ,FD t+1 ,...,FD t+9 The time series {FD} is stationary; if |Z|≥1.96, the time series {FD} is stationary. t ,FD t+1 ,...,FD t+9 It is non-stationary.

[0039] Furthermore, S32 includes:

[0040] When the error form of the abnormal optical current transformer is the ratio deviation A1, it is confirmed that the electronic unit of the abnormal optical current transformer is affected by the temperature change of the converter station.

[0041] When the error form of the abnormal optical current transformer is drift deviation A2, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the temperature change of the converter station.

[0042] When the error form of the abnormal optical current transformer is random error A3, the interference effect of the electronic unit of the abnormal optical current transformer in the strong electromagnetic environment of the converter station is confirmed.

[0043] When the error form of the abnormal optical current transformer is a fixed deviation A4, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the vibration or electromagnetic interference of the converter station.

[0044] The present invention also provides a status detection system for key components of an optical current transformer, comprising:

[0045] The status self-test module is used to synchronously collect measurement data from three optical current transformers at the same node in the converter station, and calculate the CRS, which characterizes the overall error status of the three optical current transformers at the same node. t Statistical measure and its statistical threshold CRS c Online self-detection of the overall error status of three optical current transformers at the same node;

[0046] The degradation mode determination module is used to establish the overall error state (CRS) of three optical current transformers at the same node. tTime series of statistics, calculate the fractal dimension of the time series. When the overall error states of three optical current transformers under the same node are abnormal, according to the differences in the fractal dimensions, judge the deterioration forms of the overall error states of the three optical current transformers under the same node;

[0047] A fault location determination module, configured to, according to the deterioration forms of the overall error states of three optical current transformers under the same node, in combination with a fault tree model of the deterioration characteristics of the measurement errors of the optical current transformers, reverse the abnormal influence sources of the measurement errors of the optical current transformers.

[0048] Furthermore, the state self-check module is further configured to:

[0049] S11. Collect the measurement data x of three optical current transformers under the same node of the converter station t =(I 1t , I 2t , I 3t ), where t is the sampling moment of the measurement data of the optical current transformer, and I 1t , I 2t , I 3t are respectively the current amplitude data corresponding to the three optical current transformers under the same node;

[0050] S12. Based on the measurement data x of three optical current transformers under the same node t , calculate the CRS statistic characterizing the overall error state of the three optical current transformers under the same node at this sampling moment and its statistical threshold CRS for measuring whether the error exceeds the standard c ;

[0051] S13. When the sampling moment t > n, if CRS t ≤ CRS c , the optical current transformer is in normal operation; if CRS t > CRS c , the optical current transformer operates abnormally under the interference of the abnormal influence source, indicating that there is an optical current transformer with abnormal key components among the three optical current transformers. Judge the specific abnormal optical current transformer by calculating the contribution rate of the measurement data of the three optical current transformers to the statistical characteristic quantity SPE. The optical current transformer with the largest contribution rate is considered the abnormal optical current transformer.

[0052] Even further, the S12 includes:

[0053] S121. Select the measurement data in the time period of 0 < t ≤ n, and form a data matrix Perform singular value decomposition on the covariance matrix R of the data matrix X to obtain: R = X TX / (n - 1) = [P1 P2 P3]Λ[P1 P2 P3] T

[0054] Where Λ = diag(λ1, λ2, λ3), λ1 ≥ λ2 ≥ λ3, and the vectors P1, P2, and P3 are the eigenvectors corresponding to the eigenvalues λ1, λ2, and λ3;

[0055] S122. Calculate the Hotelling T 2 statistic and the SPE statistic of the data matrix X respectively, and the calculation methods are as follows:

[0056] Where P e = [P2 P3];

[0057] S123. Calculate the expected values E(T 2 ), E(SPE) and the overall standard deviations σ(T 2 ), σ(SPE) of the statistical characteristic quantities Hotelling T 2 ) and SPE respectively;

[0058] S1244. Calculate the CRS t statistic that characterizes the overall error state of the three optical current transformers under the same node at the sampling time t (t > 0), and the calculation method is:

[0059] S125. Select the CRS t statistic that characterizes the overall error state of the three optical current transformers under the same node during the time period 0 < t ≤ n, and calculate the expected value E(CRS) and the overall standard deviation σ(CRS) of the statistical characteristic quantity CRS t ;

[0060] S126. Based on the expected value E(Q) and the overall standard deviation σ(Q), calculate the statistical threshold CRS c that characterizes whether the overall error state of the three optical current transformers under the same node exceeds the tolerance, CRS

[0061] Furthermore, the calculation process of the contribution rate of the measurement data of the three optical current transformers to the statistical characteristic quantity SPE in S13 is:

[0062]

[0063] Where J it is the contribution rate of the measurement data of the i-th optical current transformer at the sampling time t to the modulus SPE t ; x it represents the measurement data of the abnormal optical current transformer at time t; represents the projection of the measurement data of the abnormal optical current transformer at time t in the principal component subspace, where P represents the eigenvector P1 corresponding to the largest eigenvalue λ1 of the matrix R.

[0064] Furthermore, the deterioration form determination module is further configured to:

[0065] S21. Select the first n sampling times, establish a time series {CRS1, CRS2,... CRS n} of the CRS statistic, and calculate the fractal dimension FD0 of this time series;

[0066] S22. When the sampling time t > n and CRS t > CRS c , establish a time series {CRS t , CRS t+1 ,... CRS t+9} of the CRS statistic, and calculate the fractal dimension FD t ;

[0067] S23. When |FD t - FD0| / FD0 * 100% > 1% and FD t < FD0, determine that the measurement error of the abnormal optical current transformer is the ratio deviation A1 or the drift deviation A2. At this time, calculate the standard deviations S0 and S1 of the time series {CRS1, CRS2,... CRS n} and {CRS t , CRS t+1 ,... CRS t+9} respectively; when |S1 - S0| / S0 * 100% ≤ 5%, determine that the measurement error of the abnormal optical current transformer is the ratio deviation A1; when |S1 - S0| / S0 * 100% > 5%, determine that the measurement error of the abnormal optical current transformer is the drift deviation A2;

[0068] When |FD t - FD0| / FD s * 100% > 1% and FD[[ID=4�]] t > FD0, determine that the measurement error of the abnormal optical current transformer is the random error increase A3;

[0069] When |FD t - FD0| / FD0 * 100% ≤ 1%, determine that the measurement error of the abnormal optical current transformer is the fixed deviation A4.

[0070] Even further, the fault location determination module is further configured to:

[0071] S31, When sampling time t > n and CRS t >CRS c At that time, the time series of its fractal dimension {FD} is established. t ,FD t+1 ,...,FD t+9}, perform stationarity tests on data sequences with fractal dimensions;

[0072] If the data sequence of fractal dimension is stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a constant influencing factor T1, and the fault location of the abnormal optical current transformer needs to be further located.

[0073] If the data sequence of fractal dimension is non-stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a non-constant influencing factor T2, and there is no need to further locate the fault location of the abnormal optical current transformer.

[0074] S32. When the influencing factor of the measurement error of the abnormal optical current transformer is a constant influencing factor T1, it is necessary to locate the fault location of the abnormal optical current transformer based on the classification of fractal dimension and the fault tree model of the optical current transformer.

[0075] Furthermore, S31 includes:

[0076] S311, Calculate the time series {FD t ,FD t+1 ,...,FD t+9 The mean of} Through formula Construct the statistic Z, where E = 2N1N2 / 10 + 1, D = 2N1N2(2N1N2-10) / 900, and N1 is the fractal dimension of the time series less than 1. The number of elements, N² is the fractal dimension of the time series greater than 1. The number of;

[0077] S322. If |Z| < 1.96, the time series {FD} t ,FD t+1 ,...,FD t+9 The time series {FD} is stationary; if |Z|≥1.96, the time series {FD} is stationary. t ,FD t+1 ,...,FD t+9 It is non-stationary.

[0078] Furthermore, S32 includes:

[0079] When the error form of the abnormal optical current transformer is the ratio deviation A1, it is confirmed that the electronic unit of the abnormal optical current transformer is affected by the temperature change of the converter station.

[0080] When the error form of the abnormal optical current transformer is drift deviation A2, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the temperature change of the converter station.

[0081] When the error form of the abnormal optical current transformer is random error A3, the interference effect of the electronic unit of the abnormal optical current transformer in the strong electromagnetic environment of the converter station is confirmed.

[0082] When the error form of the abnormal optical current transformer is a fixed deviation A4, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the vibration or electromagnetic interference of the converter station.

[0083] The advantages of this invention are:

[0084] (1) The present invention calculates the CRS characterizing the overall error state of three optical current transformers at the same node. t Statistical measure and its statistical threshold CRS c This invention performs online self-detection of the overall error status of three optical current transformers at the same node. When an abnormality occurs in the overall error status of the three optical current transformers at the same node, the degradation mode of the overall error status of the three optical current transformers at the same node is determined based on the different fractal dimensions. Based on the degradation mode of the overall error status of the three optical current transformers at the same node, combined with the fault tree model of the degradation characteristics of optical current transformer measurement errors, the abnormal influence source of optical current transformer measurement errors can be deduced. Compared with existing technologies, this invention can accurately deduce the abnormal influence source of optical current transformer measurement errors, thus avoiding potential safety hazards.

[0085] (2) This invention, without using a standard current transformer, diagnoses and identifies the sources of abnormal measurement errors in optical current transformers solely based on the analysis of the secondary output signal of the optical current transformer. This method can support the formulation of scientific and efficient operation and maintenance strategies for optical current transformers, promote the development of the optical current transformer's condition monitoring mode from periodic passive monitoring to real-time proactive self-monitoring, and improve the equipment management level of optical current transformers.

[0086] (3) Based on the measurement data of three optical current transformers at the same node, this invention realizes the self-detection of the measurement error of the optical current transformer under the condition of no standard. According to the self-detection results of the measurement error, the characteristic form of its error degradation is autonomously identified, and the abnormal influence source of the measurement error of the optical current transformer under the condition of no power outage is realized. The diagnosis results can guide power practitioners to plan and formulate the operation and maintenance measurement of the optical current transformer in advance, thereby improving the working reliability of the optical current transformer and avoiding the operation failure of the converter station caused by the abnormal data of the optical current transformer. Attached Figure Description

[0087] Figure 1 This is a flowchart of a method for detecting the status of key components of an optical current transformer as disclosed in an embodiment of the present invention;

[0088] Figure 2 This is a flowchart of step one in a method for detecting the status of key components of an optical current transformer disclosed in an embodiment of the present invention;

[0089] Figure 3 This is a schematic diagram of the fault tree model of the key components of an optical current transformer in a method for detecting the status of key components of an optical current transformer disclosed in an embodiment of the present invention. Detailed Implementation

[0090] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0091] Example 1

[0092] like Figure 1As shown in the figure, the present invention provides a method for detecting the state of key components of an optical current transformer. Based on the correlation analysis of the measurement data of three optical current transformers under the same node of a converter station, the SPE statistic is used to characterize the overall error state of the three optical current transformers under the same node. By calculating the fractal dimension of the time series of the SPE statistic, the degradation form of the overall error state of the three optical current transformers under the same node (such as random error, fixed error, ratio error, and drift error) is judged, and this is used as a key characteristic quantity. Combining with the fault tree model of the measurement error degradation of the optical current transformer under the influence of typical factors, the abnormal influence source of the measurement error degradation of the optical current transformer is inversely calculated in real time. This embodiment realizes the autonomous diagnosis of the abnormal influence source of the measurement error of the optical current transformer under the condition of power-off. The diagnosis result can guide power industry practitioners to plan and formulate the operation and maintenance measurement of the optical current transformer in advance, so as to improve the working reliability of the optical current transformer and avoid the operation failure of the converter station caused by the data abnormality of the optical current transformer. Specifically, it includes the following steps:

[0093] S1. Synchronously collect the measurement data of three optical current transformers under the same node of the converter station, and calculate the CRS statistic and its statistical threshold CRS that characterize the overall error state of the three optical current transformers under the same node based on the data correlation analysis method, and perform on-line self-detection on the overall error state of the three optical current transformers under the same node; As t shown, the S1 includes the following steps: c Figure 2 shown,

[0094] S11. Collect the measurement data x of three optical current transformers under the same node of the converter station t =(I 1t ,I 2t ,I 3t ), where t is the sampling moment of the measurement data of the optical current transformer, and I 1t ,I 2t ,I 3t are the current amplitude data corresponding to the three optical current transformers under the same node respectively;

[0095] S12. Based on the measurement data x of the three optical current transformers under the same node t , calculate the CRS statistic that characterizes the overall error state of the three optical current transformers under the same node at this sampling moment and its statistical control limit CRS for measuring whether the error exceeds the standard; The step S12 specifically includes: c

[0096] S121. Select the measurement data in the time period of 0 < t ≤ n, and form the data matrix Perform singular value decomposition on the covariance matrix R of the data matrix X to obtain:

[0097] R = X T X / (N - 1)=[P1 P2 P3]Λ[P1 P2 P3] T

[0098] Where Λ = diag(λ1, λ2, λ3), λ1 ≥ λ2 ≥ λ3, and the vectors P1, P2, P3 are the eigenvectors corresponding to the eigenvalues λ1, λ2, λ3.

[0099] S122. Calculate the Hotelling T 2 statistic and the SPE statistic of the data matrix X respectively. The calculation methods are as follows:

[0100]

[0101] Where: P e =[P2 P3]

[0102] S123. Calculate the expected values E(T 2 ), E(SPE) and the population standard deviations σ(T 2 ), σ(SPE) of the statistical characteristic quantities Hotelling T 2 ) and SPE respectively;

[0103] Specifically, the calculation formulas for the expected values E(T 2 ), E(SPE) and the population standard deviations σ(T 2 ), σ(SPE) are as follows:

[0104]

[0105] S124. Calculate the CRS 2 statistic that characterizes the overall error state of the three optical current transformers under the same node at the sampling time t (t > 0) for the statistical characteristic quantities Hotelling T 2 ), E(SPE) and the population standard deviations σ(T 2 ), σ(SPE). The calculation method is as follows: t CRS

[0106] CRS t =(T t 2 -E(T 2 )) / σ(T 2 )+(SPE t -E(SPE)) / σ(SPE)

[0107] S125. Select the CRS that characterizes the overall error state of the three optical current transformers under the same node in the time period 0 < t ≤ n tStatistic, calculate the statistical feature quantity CRS t the expected value E(CRS) thereof and its population standard deviation σ(CRS);

[0108] Specifically, the calculation formulas for the expected value E(CRS) and its population standard deviation σ(CRS) are as follows:

[0109]

[0110]

[0111] S126. Based on the expected value E(Q) and the population standard deviation σ(Q), calculate a characteristic index CRS characterizing whether the overall error state of the three optical current transformers under the same node exceeds the tolerance c : CRS c = E(CRS) + 3σ(CRS).

[0112] It should be noted that in this embodiment, the time period of 0 < t ≤ n refers to the normal operation time period of the three optical current transformers under the same node, and n can be taken as 1000.

[0113] S13. When the sampling time t > n, perform autonomous diagnosis on the error state of the optical current transformer according to the magnitude of the CRS statistic, specifically as follows:

[0114] If CRS t ≤ CRS c , it indicates that the measurement errors of the three optical current transformers under the same node are normal at this time, and the optical current transformer is in a normal operation state;

[0115] If CRS t > CRS c , it indicates that the measurement errors of the three optical current transformers under the same node are abnormal at this time, and the optical current transformer operates abnormally under the interference of abnormal influencing sources, indicating that there is an optical current transformer with abnormal key group components among the three optical current transformers. Judge the specific abnormal optical current transformer by calculating the contribution rate of the measurement data of the three optical current transformers to the statistical feature quantity SPE. The optical current transformer with the largest contribution rate is considered to be the faulty transformer, and it is necessary to further identify the influencing source causing the abnormal operation of the optical current transformer.

[0116] In one embodiment, the calculation method for the contribution rate of the measurement data of the three optical current transformers to the statistical feature quantity SPE in the step S13 is as follows:

[0117]

[0118] In the formula: J itThe contribution rate of the measurement data of the i-th optical current transformer at the sampling time t to the modulus value SPE t ; x it represents the measurement data of the abnormal optical current transformer at time t; represents the projection of the measurement data of the abnormal optical current transformer at time t in the principal component subspace.

[0119] S2. Establish a time series of the CRS statistic characterizing the overall error state of three optical current transformers under the same node, and calculate the fractal dimension of this time series. When the overall error state of the three optical current transformers under the same node is abnormal, judge the deterioration form of the overall error state of the three optical current transformers under the same node according to the difference in the fractal dimension; the specific steps of S2 are as follows:

[0120] S21. Select the first n sampling times, establish a time series {CRS1, CRS2,... CRS n} of the CRS statistic, and calculate the fractal dimension FD0 of this time series;

[0121] S22. When the sampling time t > n, and CRS t > CRS c ), establish a time series {CRS t , CRS t+1 ,... CRS t+9} of the CRS statistic, and calculate the fractal dimension FD t ;

[0122] S23. Compare the fractal dimension FD0 with FD t , and conduct autonomous diagnosis on the error deterioration form of the optical current transformer. The specific diagnosis method is as follows:

[0123] When |FD t - FD0| / FD0 * 100% > 1%, and FD t < FD0, determine that the measurement error of the abnormal optical current transformer is a ratio deviation A1 or a drift deviation A2, and further identification is still required. Specifically, when |FD t - FD0| / FD0 * 100% > 1%, and FD t < FD0, determine that the measurement error of the abnormal optical current transformer is a ratio deviation or an offset deviation, and calculate the time series {CRS1, CRS2,... CRS n} and {CRS t , CRS t+1 ,... CRS t+9The standard deviations of the abnormal optical current transformer are S0 and S1. When |S1-S0| / S0*100%≤5%, the measurement error of the abnormal optical current transformer is determined to be the ratio deviation A1. When |S1-S0| / S0*100%>5%, the measurement error of the abnormal optical current transformer is determined to be the drift deviation A2.

[0124] When |FD t -FD0| / FD s *100%>1%, and FD t When >FD0, the measurement error of the abnormal optical current transformer is determined to be a random error increase of A3;

[0125] When |FD t -FD0| / FD0*100%≤1%, the measurement error of the abnormal optical current transformer is determined to be a fixed deviation A4.

[0126] S3. Based on the degradation pattern of the overall error state of the three optical current transformers at the same node, and combined with the fault tree model of the degradation characteristics of optical current transformer measurement errors under the influence of typical factors, the abnormal influence sources of optical current transformer measurement errors are deduced, thereby achieving autonomous identification of the fault location of the optical current transformer. S3 includes the following steps:

[0127] S31: When sampling time t > n, and CRS t >CRS c At that time, the time series of its fractal dimension {FD} is established. t ,FD t+1 ,...,FD t+9}, perform stationarity tests on data sequences with fractal dimensions;

[0128] If the data sequence of fractal dimension is stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a constant influencing factor T1, and the fault location of the abnormal optical current transformer needs to be further located.

[0129] If the data sequence of fractal dimension is non-stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a non-constant influencing factor T2, and there is no need to further locate the fault location of the abnormal optical current transformer.

[0130] In one embodiment, step S31 specifically includes:

[0131] S311, Calculate the time series {FD} t ,FD t+1 ,...,FD t+9 The data exhibits discrete characteristics, and the specific steps are as follows:

[0132] (1) Calculate the time series {FD} t ,FD t+1 ,...,FD t+9 The mean of}

[0133] (2) Compare the time series {FD} respectively t ,FD t+1 ,...,FD t+9} and mean The size, N1 is the fractal dimension of the time series less than The number of fractals, N², is the fractal dimension of the time series greater than 1. The number of;

[0134] (3) Construct the statistic Z, and calculate it as follows:

[0135]

[0136] In the formula, E = 2N1N2 / 10 + 1, D = 2N1N2(2N1N2-10) / 900.

[0137] S322, based on time series {FD t ,FD t+1 ,...,FD t+9}Analyze the discrete characteristics to determine its stationarity:

[0138] If |Z| < 1.96, the time series {FD} can be determined. t ,FD t+1 ,...,FD t+9 It is stable;

[0139] If |Z|≥1.96, the time series {FD} can be determined. t ,FD t+1 ,...,FD t+9 It is non-stationary.

[0140] S32: When the influencing factor of the abnormal optical current transformer measurement error is a constant influencing factor T1, it is necessary to classify it according to the fractal dimension and based on the fault tree model of the optical current transformer, such as... Figure 3 As shown, the fault location of the optical current transformer is determined using the following method:

[0141] (1) When the error form of the optical current transformer is the ratio deviation A1, it can be confirmed that the electronic unit of the optical current transformer is affected by the temperature change of the converter station.

[0142] (2) When the error form of the optical current transformer is drift deviation A2, it can be confirmed that the fiber optic sensing unit of the optical current transformer is affected by the temperature change of the converter station.

[0143] (3) When the error form of the optical current transformer is random error A3, the interference effect of the electronic unit of the optical current transformer in the strong electromagnetic environment of the converter station can be confirmed.

[0144] (4) When the error form of the optical current transformer is a fixed deviation A4, it can be confirmed that the optical fiber sensing unit of the optical current transformer is affected by the vibration or electromagnetic environment of the converter station.

[0145] Using the above technical solutions, this embodiment targets three optical current transformers at the same node in a converter station. Based on their output current measurement data, it first obtains the statistical characteristic quantity CRS, which characterizes the overall error state of the three optical current transformers at the same node, using mathematical statistics analysis methods. t and its statistical control limit CRS c This enables self-detection of the overall error status of three optical current transformers at the same node. Furthermore, when an anomaly occurs in the overall error status of the three optical current transformers at the same node, the CRS is calculated. t The fractal dimension of the statistical time series is used to autonomously identify the degradation characteristics of the overall error state of three optical current transformers at the same node. Based on the judgment of fractal dimension stability and its fault tree model, the abnormal causes and fault locations of abnormal optical current transformers are autonomously identified to guide the operation and maintenance strategies of optical current transformers. The self-detection method for the operating status of key components of optical current transformers proposed in this embodiment is applicable to various types of optical current transformers with voltage levels of 800kV and above.

[0146] Example 2

[0147] Based on Embodiment 1, Embodiment 2 of the present invention also provides a status detection system for key components of an optical current transformer, comprising:

[0148] The status self-test module is used to synchronously collect measurement data from three optical current transformers at the same node in the converter station, and calculate the CRS, which characterizes the overall error status of the three optical current transformers at the same node. t Statistical measure and its statistical threshold CRS c Online self-detection of the overall error status of three optical current transformers at the same node;

[0149] The degradation mode determination module is used to establish the overall error state (CRS) of three optical current transformers at the same node. t The time series of statistics is calculated, and the fractal dimension of the time series is determined. When the overall error state of three optical current transformers at the same node is abnormal, the deterioration mode of the overall error state of the three optical current transformers at the same node is determined according to the different fractal dimensions.

[0150] A fault location determination module, which is used to inversely deduce the abnormal influence source of the measurement error of the optical current transformer according to the deterioration form of the overall error state of three optical current transformers under the same node, and in combination with the fault tree model of the deterioration characteristics of the measurement error of the optical current transformer, so as to realize the autonomous identification of the fault location of the optical current transformer.

[0151] Specifically, the status self-check module is further used for:

[0152] S11. Collect the measurement data x of three optical current transformers under the same node of the converter station t =(I 1t , I 2t , I 3t ), where t is the sampling time of the measurement data of the optical current transformer, and I 1t , I 2t , I 3t are the current amplitude data corresponding to three optical current transformers under the same node respectively;

[0153] S12. Based on the measurement data x of three optical current transformers under the same node t , calculate the CRS statistic representing the overall error state of three optical current transformers under the same node at this sampling time and its statistical threshold CRS for measuring whether the error exceeds the standard c ;

[0154] S13. When the sampling time t > n, if CRS t ≤ CRS c , the optical current transformer is in normal operation; if CRS t > CRS c , the optical current transformer operates abnormally under the interference of the abnormal influence source, indicating that there is an optical current transformer with abnormal key components among the three optical current transformers. Judge the specific abnormal optical current transformer by calculating the contribution rate of the measurement data of the three optical current transformers to the statistical characteristic quantity SPE, and the optical current transformer with the largest contribution rate is considered as the abnormal optical current transformer.

[0155] More specifically, the S12 includes:

[0156] S121. Select the measurement data in the time period of 0 < t ≤ n, and form a data matrix Perform singular value decomposition on the covariance matrix R of the data matrix X to obtain: R = X T X / (n - 1) = [P1 P2 P3]Λ[P1 P2 P3] T

[0157] Where, Λ = diag(λ1, λ2, λ3), λ1 ≥ λ2 ≥ λ3, Λ represents a diagonal matrix with respect to the eigenvalues λ1, λ2, λ3, diag represents a diagonal matrix, λ1, λ2, λ3 represent the eigenvalues of matrix R, and the vectors P1, P2, P3 are the eigenvectors corresponding to the eigenvalues λ1, λ2, λ3;

[0158] S122. Calculate the Hotelling T 2 statistic and the SPE statistic of the data matrix X respectively, and the calculation methods are as follows:

[0159] Where, P e = [P2 P3];

[0160] S123. Calculate the expected values E(T 2 ), E(SPE) and the overall standard deviations σ(T 2 ), σ(SPE) of the statistical characteristic quantities Hotelling T 2 and SPE respectively;

[0161] S124. Calculate the CRS t statistic characterizing the overall error state of the three optical current transformers under the same node at the sampling time t (t > 0), and the calculation method is:

[0162] S125. Select the CRS t statistic characterizing the overall error state of the three optical current transformers under the same node in the time period 0 < t ≤ n, and calculate the expected value E(CRS) and the overall standard deviation σ(CRS) of the statistical characteristic quantity CRS t ;

[0163] S126. Based on the expected value E(Q) and the overall standard deviation σ(Q), calculate the statistical threshold CRS c characterizing whether the overall error state of the three optical current transformers under the same node is out of tolerance, CRS

[0164] More specifically, the calculation process of the contribution rate of the measurement data of the three optical current transformers to the statistical characteristic quantity SPE in S13 is:

[0165]

[0166] Where, J it is the contribution rate of the measurement data of the i-th optical current transformer at the sampling time t to the modulus SPE t ; x it represents the measurement data of the abnormal optical current transformer at time t; It represents the projection of the measurement data of the abnormal optical current transformer at time t in the principal component subspace, where P represents the eigenvector P1 corresponding to the largest eigenvalue λ1 of the matrix R.

[0167] Specifically, the degradation form determination module is further configured to:

[0168] S21. Select the previous n sampling moments, establish the time series {CRS1, CRS2,... CRS n} of the CRS statistic, and calculate the fractal dimension FD0 of this time series;

[0169] S22. When the sampling moment t > n and CRS t > CRS c ), establish the time series {CRS t , CRS t+1 ,... CRS t+9} of the CRS statistic, and calculate the fractal dimension FD t ;

[0170] S23. When |FD t - FD0| / FD0 * 100% > 1% and FD t < FD0, determine that the measurement error of the abnormal optical current transformer is the ratio deviation A1 or the drift deviation A2. At this time, calculate the standard deviations S0 and S1 of the time series {CRS1, CRS2,... CRS n} and {CRS t , CRS t+1 ,... CRS t+9} respectively; when |S1 - S0| / S0 * 100% ≤ 5%, determine that the measurement error of the abnormal optical current transformer is the ratio deviation A1; when |S1 - S0| / S0 * 100% > 5%, determine that the measurement error of the abnormal optical current transformer is the drift deviation A2;

[0171] When |FD t - FD0| / FD s * 100% > 1% and FD t > FD0, determine that the measurement error of the abnormal optical current transformer is the random error increase A3;

[0172] When |FD t - FD0| / FD0 * 100% ≤ 1%, determine that the measurement error of the abnormal optical current transformer is the fixed deviation A4.

[0173] More specifically, the fault location determination module is further configured to:

[0174] S31. When the sampling moment t > n and CRSt >CRS c At that time, the time series of its fractal dimension {FD} is established. t ,FD t+1 ,...,FD t+9}, perform stationarity tests on data sequences with fractal dimensions;

[0175] If the data sequence of fractal dimension is stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a constant influencing factor T1, and the fault location of the abnormal optical current transformer needs to be further located.

[0176] If the data sequence of fractal dimension is non-stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a non-constant influencing factor T2, and there is no need to further locate the fault location of the abnormal optical current transformer.

[0177] S32. When the influencing factor of the measurement error of the abnormal optical current transformer is a constant influencing factor T1, it is necessary to locate the fault location of the abnormal optical current transformer based on the classification of fractal dimension and the fault tree model of the optical current transformer.

[0178] More specifically, S31 includes:

[0179] S311, Calculate the time series {FD t ,FD t+1 ,...,FD t+9 The mean of} Through formula Construct the statistic Z, where E = 2N1N2 / 10 + 1, D = 2N1N2(2N1N2-10) / 900, and N1 is the fractal dimension of the time series less than 1. The number of elements, N² is the fractal dimension of the time series greater than 1. The number of;

[0180] S322. If |Z| < 1.96, the time series {FD} t ,FD t+1 ,...,FD t+9 The time series {FD} is stationary; if |Z|≥1.96, the time series {FD} is stationary. t ,FD t+1 ,...,FD t+9 It is non-stationary.

[0181] More specifically, S32 includes:

[0182] When the error form of the abnormal optical current transformer is the ratio deviation A1, it is confirmed that the electronic unit of the abnormal optical current transformer is affected by the temperature change of the converter station.

[0183] When the error form of the abnormal optical current transformer is drift deviation A2, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the temperature change of the converter station.

[0184] When the error form of the abnormal optical current transformer is random error A3, the interference effect of the electronic unit of the abnormal optical current transformer in the strong electromagnetic environment of the converter station is confirmed.

[0185] When the error form of the abnormal optical current transformer is a fixed deviation A4, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the vibration or electromagnetic interference of the converter station.

[0186] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An optical current transformer key component status detection method, characterized by, include: Step 1: Synchronously collect measurement data from three optical current transformers at the same node of the converter station, and calculate the overall error state of the three optical current transformers at the same node. Statistics and their statistical thresholds Online self-detection of the overall error status of three optical current transformers at the same node; Step 2: Establish a characterization of the overall error state of the three optical current transformers at the same node. The time series of statistics is calculated, and the fractal dimension of the time series is determined. When the overall error state of the three optical current transformers at the same node becomes abnormal, the degradation mode of the overall error state of the three optical current transformers at the same node is determined based on the different fractal dimensions. Step two includes: S21、selecting the first n sampling time, establishing the time series of CRS statistics , calculating the fractal dimension of the time series ; S22, when the sampling time t > n, and a time series of the CRS statistics is established the fractal dimension of the time series is calculated ; S23, when ,and When the measurement error of the abnormal optical current transformer is determined to be either the ratio deviation A1 or the drift deviation A2, the time series is calculated respectively. , The standard deviation is , ;when When, the measurement error of the abnormal optical current transformer is determined to be the ratio deviation A1; when At that time, the measurement error of the abnormal optical current transformer is determined to be the drift deviation A2; When , and , the measurement error of the abnormal optical current transformer is determined as a random error increase A3. When , the measurement error of the abnormal optical current transformer is determined as a fixed deviation A4; Step 3: Based on the overall error state degradation of the three optical current transformers at the same node, and combined with the fault tree model of the measurement error degradation characteristics of the optical current transformers, the abnormal influence sources of the optical current transformer measurement error are deduced.

2. The method of claim 1, wherein the method further comprises: Step one includes: S11, collect the measurement data of three optical current transformers under the same node of the converter station t is the sampling time of the optical current transformer measurement data, are the current amplitude data corresponding to the three optical current transformers under the same node, respectively S12、based on the same node under three optical current transformer measurement data , calculate the sampling time representing the same node under three optical current transformer overall error state CRS statistical amount and whether the statistical threshold of its measurement error is out of tolerance ; S13, when sampling time t > n At that time, if The optical current transformer is in normal operating condition; if The optical current transformers malfunctioned under interference from abnormal sources, indicating that a key component of one of the three optical current transformers was faulty. This was determined by calculating the statistical characteristics of the measurement data from the three optical current transformers. SPE The contribution rate is used to determine the specific abnormal optical current transformer. The optical current transformer with the largest contribution rate is considered to be an abnormal optical current transformer.

3. The method of claim 2, wherein the method further comprises: S12 includes: S121, select 0 t ≤ n measurement data of the time period, the data matrix , the covariance matrix of the data matrix X singular value decomposition to obtain: ​ In the formula, , , Represents the relationship with eigenvalues diagonal matrix, Represents a diagonal matrix. Representation matrix R eigenvalues, vectors , , Eigenvalues Corresponding feature vectors; S122, respectively calculate the data matrix X of HotellingT 2 statistical quantities and SPE statistical quantities, the calculation method is as follows: , wherein ; S123. Calculate the statistical characteristic quantities respectively. HotellingT 2 , SPE Expected value , and its overall standard deviation , ; S124, calculating sampling time t , t >0, representing the overall error state of three optical current transformers under the same node statistical quantity, the calculation method is: S125, Select 0< t ≤ n The time period characterizes the overall error state of the three optical current transformers at the same node. Statistical measure, calculating the statistical characteristic measure Expected value and its overall standard deviation ; S126、based on the expected value and the overall standard deviation , calculate a statistical threshold value representing whether the overall error state of the three optical current transformers under the same node is out of tolerance .

4. The method of claim 3, wherein the method further comprises: The measurement data from the three optical current transformers in S13 are statistical characteristic quantities. SPE The calculation process for the contribution rate is as follows: In the formula, The measurement data of the i-th optical current transformer at sampling time t is compared with the modulus value. The contribution rate; This represents the measurement data of the abnormal optical current transformer at time t; = , represents the projection of the measurement data of the abnormal optical current transformer at time t into the principal element subspace, where, Represents the largest eigenvalue of matrix R Corresponding feature vector .

5. The method of claim 1, wherein, Step three includes: S31, when sampling time t > n and , the time series of the fractal dimension is established , , the data sequence of the fractal dimension is stationary If the data sequence of fractal dimension is stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a constant influencing factor T1, and the fault location of the abnormal optical current transformer needs to be further located. If the data sequence of fractal dimension is non-stationary, then the influencing factor of the measurement error of the abnormal optical current transformer is determined to be a non-constant influencing factor T2, and there is no need to further locate the fault location of the abnormal optical current transformer. S32. When the influencing factor of the measurement error of the abnormal optical current transformer is a constant influencing factor T1, it is necessary to locate the fault location of the abnormal optical current transformer based on the classification of fractal dimension and the fault tree model of the optical current transformer.

6. The method of claim 5, wherein the method further comprises: S31 includes: S311, calculating the mean value of the time series , } of the time series ; constructing a statistic Z by the formula , where , , is the number of points in the time series with a fractal dimension less than , is the number of points in the time series with a fractal dimension greater than ; S322、if , the time series { , } is stationary; if , the time series { , } is non-stationary.

7. The method of claim 5, wherein the method further comprises: S32 includes: When the error form of the abnormal optical current transformer is the ratio deviation A1, it is confirmed that the electronic unit of the abnormal optical current transformer is affected by the temperature change of the converter station. When the error form of the abnormal optical current transformer is drift deviation A2, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the temperature change of the converter station. When the error form of the abnormal optical current transformer is random error A3, the interference effect of the electronic unit of the abnormal optical current transformer in the strong electromagnetic environment of the converter station is confirmed. When the error form of the abnormal optical current transformer is a fixed deviation A4, it is confirmed that the fiber optic sensing unit of the abnormal optical current transformer is affected by the vibration or electromagnetic interference of the converter station.

8. An optical current transformer key component status detection system, characterized by, include: The status self-test module is used to synchronously collect measurement data from three optical current transformers at the same node in the converter station, and calculate the overall error status of the three optical current transformers at the same node. Statistics and their statistical thresholds Online self-detection of the overall error status of three optical current transformers at the same node; The degradation mode determination module is used to establish the overall error state of three optical current transformers at the same node. The time series of statistics is calculated, and the fractal dimension of the time series is determined. When the overall error state of three optical current transformers at the same node becomes abnormal, the degradation mode of the overall error state of the three optical current transformers at the same node is determined based on the different fractal dimensions. The degradation mode determination module is also used for: S21, select the first n sampling time, establish the time series of CRS statistics , calculate the fractal dimension of the time series ; S22, when the sampling time t > n, and a time series of the CRS statistics is established the fractal dimension of the time series is calculated ; S23, when ,and When the measurement error of the abnormal optical current transformer is determined to be either the ratio deviation A1 or the drift deviation A2, the time series is calculated respectively. , The standard deviation is , ;when When, the measurement error of the abnormal optical current transformer is determined to be the ratio deviation A1; when At that time, the measurement error of the abnormal optical current transformer is determined to be the drift deviation A2; When , and , the measurement error of the abnormal optical current transformer is determined as a random error increase A3. When , the measurement error of the abnormal optical current transformer is determined as a fixed deviation A4; The fault location determination module is used to infer the abnormal influence source of the optical current transformer measurement error based on the overall error state degradation of the three optical current transformers at the same node and the fault tree model of the measurement error degradation characteristics of the optical current transformer.

9. An optical current transformer key component status detection system according to claim 8, wherein, The status self-test module is also used for: S11, collect the measurement data of three optical current transformers under the same node of the converter station t is the sampling time of the optical current transformer measurement data, are the current amplitude data corresponding to the three optical current transformers under the same node, respectively S12、based on the same node under three optical current transformer measurement data , calculate the sampling time representing the same node under three optical current transformer overall error state CRS statistical amount and whether its measure error is out of tolerance statistical threshold ; S13, when sampling time t > n At that time, if The optical current transformer is in normal operating condition; if The optical current transformers malfunctioned under interference from abnormal sources, indicating that a key component of one of the three optical current transformers was faulty. This was determined by calculating the statistical characteristics of the measurement data from the three optical current transformers. SPE The contribution rate is used to determine the specific abnormal optical current transformer. The optical current transformer with the largest contribution rate is considered to be an abnormal optical current transformer.

Images