An online quantitative analysis method and system for error state of an optical fiber current transformer

Abstract

The application discloses an optical fiber current transformer error state online quantitative analysis method and system, and the method comprises the following steps: obtaining single-body reliability by calculating two-by-two coupling difference degree, and obtaining recursive reliability by fusing with historical information to dynamically screen reliable reference subgroups; taking the channel with the maximum recursive reliability as a temporary reference to perform coarse alignment, and then using the reliable subgroups to construct a group reference waveform to perform fine alignment and output synchronization deviation; extracting amplitude deviation, waveform residual error and correlation decline according to the difference between the final waveform and the group reference waveform to jointly constitute four types of error state components; and performing normalization based on a dynamic division working condition baseline to obtain a comprehensive health index, and calculating a weighted continuous abnormality proportion, a trend correction slope and a comprehensive degradation score to realize short-time abnormality alarm and long-term degradation trend evaluation. The application does not need an external standard device, and has the advantages of strong robustness, good interpretability, working condition self-adaptation and the like.

Description

Technical Field

[0001] This invention belongs to the field of power system measurement and condition assessment technology, and particularly relates to an online quantitative analysis method and system for the error condition of fiber optic current transformers. Background Technology

[0002] Fiber optic current transformers (FOCTs) have advantages such as good insulation performance, high bandwidth, strong resistance to electromagnetic interference, and suitability for digital measurement. They have been widely used in converter stations, online monitoring of main power grid equipment, and high-voltage measurement. Configuring multiple FOCTs at the same measuring point to form redundant measurement is a common engineering approach to improve operational reliability and facilitate fault diagnosis.

[0003] With long-term operation, the optical link, electronic conditioning link, temperature compensation unit, and synchronization and sampling link of the FOCT may experience drift, mismatch, or local degradation, which may manifest as amplitude deviation, time base loss, waveform distortion, and increased noise. Such changes may not immediately cause obvious faults, but they will gradually weaken the reliability of measurements and affect the condition awareness and operation and maintenance decisions of the converter station.

[0004] Currently, common engineering practices mainly include offline verification, simple threshold alarms, and error monitoring based on single-unit output. Offline verification requires power outages or disassembly for inspection, and cannot reflect the dynamic deterioration process during operation; the single-unit threshold method is easily affected by primary current fluctuations, changes in operating conditions, and local interference, making it difficult to achieve continuous, quantitative, and online evaluation of individual FOCTs without external standard current transformers.

[0005] The existing technology has the following main drawbacks: Reference construction is susceptible to contamination. Existing multi-machine consensus methods typically assume that all channels are trustworthy and directly participate in the reference construction. When there are synchronization anomalies, amplitude drift, or waveform distortion in some channels, the reference waveform itself will be contaminated, leading to distorted subsequent evaluation results.

[0006] The time base problem is coupled with the measurement problem. Existing methods often only perform a single alignment or do not retain synchronization deviation information separately, which can easily mask the time base out-of-sync after alignment, making it difficult to distinguish between synchronization chain anomalies and measurement chain anomalies.

[0007] Fixed thresholds have poor adaptability. Under different measuring points, different current loads, different temperatures, and different operating modes, the normal fluctuation range of consistency indicators is not the same, and fixed thresholds are prone to false alarms or missed alarms.

[0008] The evaluation results lack interpretability. Some schemes can only provide "normal / abnormal" or a single health score, making it difficult to simultaneously quantify different error state components such as synchronization deviation, amplitude deviation, waveform residual, and correlation decline.

[0009] There is a lack of long-term degradation assessment capabilities. Single-time-window assessments are insufficient to reflect the continuous deterioration trend of equipment, which is not conducive to forming early warning and operation and maintenance decisions based on condition-based maintenance.

[0010] Therefore, there is an urgent need in this field for a method and system that can quantitatively assess the synchronization anomalies, amplitude drift, waveform distortion, and long-term degradation of a single FOCT online by utilizing the group consistency and operating condition adaptive baseline of multiple FOCTs at the same measurement point without relying on standard current transformers. Summary of the Invention

[0011] This invention aims to provide an online quantitative analysis method and system for the error status of fiber optic current transformers (FOCTs). Under the condition of no external standard transformer, it achieves online identification and graded evaluation of synchronization anomalies, amplitude drift, waveform distortion and overall degradation of a single FOCT through reliable group recursive screening, two-level time alignment, consistency feature extraction, dynamic working condition baseline update, error status quantification and trend evaluation.

[0012] In a first aspect, the present invention provides an online quantitative analysis method for the error status of an optical fiber current transformer, applicable to redundant current measurement scenarios of n optical fiber current transformers at the same measurement point, wherein n≥3, the method comprising: Acquire the current waveform data of each current transformer and perform preprocessing, then filter out the effective time window based on the effective value threshold; Within the effective time window, the weighted coupling difference between every two current transformers is calculated. Based on the weighted coupling difference, the individual reliability of each current transformer in the current time window is calculated. Then, the individual reliability is weighted and fused with the current transformer's own historical recursive reliability to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically selected. Select the current transformer with the highest recursive confidence from all transformers as the temporary reference transformer, and perform the first time-delay alignment of the current waveforms of each transformer relative to the temporary reference transformer to obtain the coarse alignment waveform. A group reference waveform is constructed using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup. The coarse alignment waveforms of each mutual inductor are then re-aligned with the group reference waveform to obtain a fine alignment correction. The sum of the delay from the first time-delay alignment and the fine alignment correction from the second time-delay alignment is taken as the total alignment amount. The total alignment amount is output as the synchronization deviation. The final alignment waveform is obtained based on the fine alignment correction. Based on the difference between the final aligned waveform and the group reference waveform, the amplitude deviation, waveform residual and correlation decrease value are extracted, and together with the synchronization deviation, they constitute the error state component. Based on the dynamic working condition baseline, the error state components are normalized respectively to obtain the normalized deviation of each current transformer, and the normalized deviation of each current transformer is weighted and summed to obtain the comprehensive health index of each current transformer. Based on the historical time window series of the comprehensive health index of the current transformer, the weighted continuous abnormality ratio, trend correction slope and comprehensive degradation score are calculated to perform short-term abnormality alarm and long-term degradation trend assessment for a single current transformer.

[0013] Secondly, this invention provides an online quantitative analysis system for the error status of fiber optic current transformers, applicable to redundant current measurement scenarios of n fiber optic current transformers at the same measurement point, wherein n≥3, and the system includes: The acquisition module is configured to acquire the current waveform data of each current transformer and perform preprocessing, and filter out the effective time window based on the effective value threshold. The filtering module is configured to calculate the weighted coupling difference between every two current transformers within the effective time window, calculate the individual reliability of each current transformer in the current time window based on the weighted coupling difference, and then perform weighted fusion with the current transformer's own historical recursive reliability to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically filtered out. The alignment module is configured to select the one with the highest recursive confidence from all current transformers as a temporary reference current transformer, and perform the first time-delay alignment of the current waveforms of each current transformer relative to the temporary reference current transformer to obtain a coarse alignment waveform. The output module is configured to construct a group reference waveform using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup, and perform a second time delay alignment on the coarse alignment waveforms of each mutual inductor relative to the group reference waveform to obtain a fine alignment correction amount. The sum of the time delay of the first time delay alignment and the fine alignment correction amount of the second time delay alignment is taken as the total alignment amount. The total alignment amount is output as the synchronization deviation, and the final alignment waveform is obtained based on the fine alignment correction amount. The extraction module is configured to extract amplitude deviation, waveform residual and correlation decrease value based on the difference between the final aligned waveform and the group reference waveform, and together with the synchronization deviation, they constitute the error state component. The normalization module is configured to normalize the error state components based on the dynamic working condition baseline to obtain the normalized deviation of each current transformer, and then weighted summation of the normalized deviation of each current transformer to obtain the comprehensive health index of each current transformer. The analysis module is configured to calculate the weighted continuous abnormality ratio, trend correction slope, and comprehensive degradation score based on the historical time window sequence of the comprehensive health index of the current transformer, so as to perform short-term abnormality alarms and long-term degradation trend assessments for a single current transformer.

[0014] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the online quantitative analysis method for the error state of an optical fiber current transformer according to any embodiment of the present invention.

[0015] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the steps of the online quantitative analysis method for the error state of an optical fiber current transformer according to any embodiment of the present invention.

[0016] The online quantitative analysis method and system for the error status of fiber optic current transformers disclosed in this application have the following advantages: The group consistency of redundant FOCTs at the same measurement point can be used to output quantifiable error state quantities, which has significant value for online applications. By using a screening chain of "pairwise coupling difference degree → individual confidence degree → recursive confidence degree → confidence subgroup", the stability of the confidence reference construction is improved, and it is more robust than a simple threshold elimination scheme. Through a two-stage time base correction process of "temporary reference coarse alignment + group reference fine alignment", the total alignment amount is retained as independent synchronization health information, which makes it easier to distinguish between synchronization anomalies and measurement anomalies. The four categories of quantification indicators—synchronization deviation, amplitude deviation, waveform residual, and correlation reduction—enhance the interpretability of anomalies and facilitate the differentiation of different fault mechanisms and maintenance handling methods. By using a dynamic working condition baseline center and scale update mechanism, the problem of insufficient adaptability of fixed thresholds to different loads, temperatures and measurement points is reduced. By combining the weighted proportion of persistent anomalies, the trend correction slope, and the comprehensive degradation score, a combination of single-time-window judgment and long-term trend early warning is achieved, which is suitable for closed-loop engineering applications. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart of an online quantitative analysis method for the error state of an optical fiber current transformer provided in an embodiment of the present invention; Figure 2This is a structural block diagram of an online quantitative analysis system for the error status of an optical fiber current transformer, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] 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 with reference to the accompanying drawings. 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.

[0020] Please see Figure 1 The flowchart illustrates an online quantitative analysis method for the error state of an optical fiber current transformer according to this application.

[0021] like Figure 1 As shown, the online quantitative analysis method for the error status of fiber optic current transformers specifically includes the following steps: Step S101: Obtain the current waveform data of each current transformer and perform preprocessing; then filter out the effective time window based on the effective value threshold.

[0022] In this step, raw current waveform data from n fiber optic current transformers (FOCTs) at the same measurement point are acquired using a high-precision synchronous sampling device. To ensure the accuracy of subsequent time delay alignment, the sampling clocks of all transformers should be synchronized, and the sampling frequency should be... Typically set to 10 kHz to 100 kHz (e.g., 25.6 kHz is preferred in engineering), the sampled data for each channel is stored in time-stamp alignment.

[0023] Let the original current signal output by the i-th current transformer within the time window p be... t is the time variable. For ease of digital processing, a discrete sampling sequence is actually obtained. , , This represents the total number of sampling points within the time window. The length of the basic analysis time window can be 1 second or cover an integer number of power frequency cycles in engineering. Since the FOCT output may contain DC bias (caused by preamplifier zero drift, analog-to-digital converter bias, etc.), DC removal processing is required for each signal. The specific method for DC removal is as follows: calculate the arithmetic mean of the original signal within the current time window, and then subtract this mean from each sampling point to obtain the zero-mean AC component. For discrete sequences, the calculation formula is: , , In the formula, Let be the discrete current value after DC removal from the k-th current transformer. This is the arithmetic mean of the original current waveform of the k-th current transformer within the current time window; When the data is expressed as a continuous waveform, it can be written as: , In the formula, For time t from 0 to definite integral, The original current waveform is in continuous form; To suppress the impact of high-frequency noise (such as switching power supply interference and electromagnetic radiation coupling noise) on subsequent correlation and time delay estimation, a zero-phase digital filter is used to perform low-pass filtering on the DC-decoupled signal. Zero-phase filtering can be achieved through a combination of forward and reverse filtering (e.g., using the `filtfilt` function in Matlab), avoiding the introduction of phase delay. The filter cutoff frequency can be set according to the power frequency signal frequency, typically 5 to 10 times the power frequency (e.g., for a 50Hz system, the cutoff frequency is set to 250Hz–500Hz).

[0024] Calculate the effective value of the filtered signal: , In the formula, Let be the effective current value of the k-th current transformer within the current time window. The continuous waveform of the k-th current transformer after zero-phase low-pass filtering; Set minimum effective current threshold Typically, 0.5% to 2% of the rated current of the current transformer is taken (for example, when the rated current is 1000A, take...). =10A). If all channels within the current time window All less than If the primary current is too small and the signal-to-noise ratio is too low, the health evaluation result for this time window will not be output, and subsequent steps will be skipped directly; otherwise, the time window will be marked as an "effective time window", and the filtered waveform will be used as the input for subsequent steps.

[0025] Step S102: Within the effective time window, calculate the weighted coupling difference between every two current transformers, calculate the individual reliability of each current transformer in the current time window based on the weighted coupling difference, and then perform weighted fusion with the historical recursive reliability of the current transformer to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, dynamically select a reliable reference subgroup.

[0026] In this step, the current amplitude of the i-th current transformer within the p-th effective time window is used as the basis for the calculation. The current amplitude of the j-th current transformer within the p-th time window Waveform correlation coefficient and optimal alignment delay The weighted coupling difference between the i-th and j-th current transformers is calculated using the following expression: , In the formula, For reference amplitude, Let be the time delay normalization constant. These are weighted coefficients, and their sum is 1. It is a very small positive number; Among them, waveform correlation coefficient The Pearson correlation coefficient between the waveforms of the i-th and j-th current transformers is given by the optimal alignment delay. The time shift required for the two waveforms to achieve maximum correlation is obtained by the peak position of the cross-correlation function.

[0027] The individual reliability of the i-th current transformer within the p-th effective time window is calculated based on the weighted coupling difference. The expression is: , In the formula, This represents the total number of current transformers. It is an exponential function with the natural constant e as its base. This represents the summation of all j except i; The current individual confidence level is weighted and fused with the recursive confidence level of the previous effective time window of the current transformer using a time memory factor, resulting in the recursive confidence level of the current effective time window, expressed as: , In the formula, Let i be the recursive reliability of the i-th current transformer within the p-th valid time window. As a time memory factor, , The historical recursive reliability of the i-th current transformer within the p-1-th effective time window; Calculate the mean recursive reliability of all current transformers within the current valid time window. and standard deviation And determine the trusted reference subgroup according to the preset filtering rules. The expression for the filtering rule is: , In the formula, This is the adjustment coefficient.

[0028] Step S103: Select the current transformer with the highest recursive confidence from all current transformers as a temporary reference current transformer, and perform the first time delay alignment of the current waveform of each current transformer relative to the temporary reference current transformer to obtain the coarse alignment waveform.

[0029] In this step, the current transformer with the highest recursive confidence level is selected from all current transformers as the temporary reference current transformer. The numbering of the temporary reference current transformer satisfies the following condition: , In the formula, Let i be the recursive reliability of the i-th current transformer within the p-th valid time window. This is the number of the temporary reference transformer. In order to make The value of i when the maximum value is reached; Calculate the coarse alignment delay of the i-th current transformer relative to the temporary reference current transformer. The expression for the coarse alignment delay is: , , In the formula, The coarse alignment delay of the i-th current transformer relative to the temporary reference current transformer is given by: For time delay variables, After removing DC from the i-th current transformer, Current waveform at time t, For the temporary reference transformer in the p-th effective time window Waveform at time, After removing DC power from the kth current transformer, Current waveform at time t, It is a cross-correlation function; The coarse-aligned waveform is obtained by shifting the coarse-aligned delay waveform, and the expression is: , In the formula, This is the coarse alignment waveform of the i-th current transformer after the first time-delay alignment. For time variables, Let i be the waveform function after DC removal from the i-th current transformer. This indicates that the original waveform is shifted along the time axis. ,when The waveform shifts to the left when it is positive. When the value is negative, the waveform shifts to the right.

[0030] Step S104: Construct a group reference waveform using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup, and perform a second time delay alignment on the coarse alignment waveforms of each mutual inductor relative to the group reference waveform to obtain a fine alignment correction amount. The sum of the time delay of the first time delay alignment and the fine alignment correction amount of the second time delay alignment is taken as the total alignment amount. The total alignment amount is output as the synchronization deviation, and the final alignment waveform is obtained based on the fine alignment correction amount.

[0031] In this step, the trusted reference subgroup The coarse-aligned waveform is used to construct a group reference waveform by taking the median point by point. The expression is: , In the formula, For group reference waveform, This is the coarse alignment waveform of the i-th current transformer after the first time-delay alignment. To obtain the median, that is, for each fixed time point t, take all... corresponding The median; The fine alignment correction for the i-th current transformer relative to the group reference waveform is calculated using the following expression: , In the formula, Let be the fine alignment correction amount of the i-th current transformer relative to the group reference waveform. For the group reference waveform in time shift The waveform after that, For time delay variables, It is a cross-correlation function; The total alignment amount is obtained by adding the coarse alignment delay and the fine alignment correction. The total alignment amount is output as the synchronization deviation, and the final alignment waveform is obtained based on the fine alignment correction amount, expressed as: , In the formula, This represents the final aligned waveform of the i-th current transformer after two steps of coarse and fine alignment correction within the p-th effective time window. For time variables, Let i be the coarse alignment waveform function for the i-th mutual inductor. This indicates that the coarse-aligned waveform will be shifted on the time axis. ,when The waveform shifts to the left when it is positive. When the value is negative, the waveform shifts to the right.

[0032] Step S105: Based on the difference between the final aligned waveform and the group reference waveform, extract the amplitude deviation, waveform residual and correlation decrease value, which together with the synchronization deviation constitute the error state component.

[0033] In this step, the expression for calculating the amplitude deviation is: , , , In the formula, For amplitude deviation, This refers to the relative amplitude deviation. The effective value of the group reference waveform, To find the root mean square value; The expression for calculating the waveform residual is as follows: , In the formula, For waveform residuals; The expression for calculating the correlation decrease value is: , In the formula, This represents the correlation decrease value. It is a cross-correlation function; The expression for calculating the synchronization deviation is: , In the formula, This is a synchronization deviation.

[0034] Step S106: Based on the dynamic working condition baseline, the error state components are normalized to obtain the normalized deviation of each current transformer, and the normalized deviation of each current transformer is weighted and summed to obtain the comprehensive health index of each current transformer.

[0035] In this step, the operating condition category for the current time window is established. , This is the working condition mapping function. For ambient temperature, For the load range, The effective value of the group reference waveform; For the j-th type of error state component, the baseline center and baseline scale are updated respectively, as expressed by: , , In the formula, , These are the center estimate and scale estimate, respectively, obtained from historical health samples under the current operating condition category. , To update the coefficients, and the values ​​range from 0 to 1, , Let p be the baseline center and baseline scale for the p-th effective time window. , The baseline center and baseline scale for the (p-1)th effective time window; Calculate the normalized deviation of the j-th type of error state component of the i-th current transformer within the p-th time window based on the baseline center and baseline scale. The expression is: , In the formula, Let be the original quantity of the j-th type of error state component of the i-th current transformer. It is a very small positive number.

[0036] Step S107: Based on the historical time window sequence of the comprehensive health index of the current transformer, calculate the weighted continuous abnormality ratio, trend correction slope and comprehensive degradation score, which are used to perform short-term abnormality alarm and long-term degradation trend assessment for a single current transformer.

[0037] In this step, the expression for calculating the weighted proportion of persistent anomalies is: , In the formula, Let be the weighted continuous anomaly ratio of the i-th current transformer over the most recent N time windows. To calculate the window length, This refers to the time window number. The comprehensive health index of the i-th current transformer in the r-th time window. Forgetting factor, As the warning threshold, It is an indicator function; The expression for calculating the trend correction slope is: , , , In the formula, Let be the trend correction slope of the i-th current transformer over the most recent N time windows. The arithmetic mean of the time sequence numbers. Let be the arithmetic mean of the comprehensive health index of the i-th current transformer over the most recent N time windows. This is the preset fluctuation correction coefficient. The comprehensive health index of the i-th current transformer in the (r-1)-th time window; The expression for calculating the comprehensive degradation score based on the weighted persistent abnormality ratio and the trend correction slope is as follows: , In the formula, This represents the overall degradation score of the i-th current transformer over the most recent N time windows. Let be the median of the comprehensive health index of the i-th current transformer over the most recent N time windows. , , These are the preset weighting coefficients.

[0038] In summary, the method of this application, for redundant measurement scenarios of multiple current transformers at the same measuring point, obtains the individual reliability by calculating the pairwise coupling difference degree under the condition of no external standard current transformer, and obtains the recursive reliability by fusing with historical information, dynamically selecting a reliable reference subgroup; coarse alignment is performed using the channel with the highest recursive reliability as a temporary reference, and then fine alignment is performed by constructing a group reference waveform using the reliable subgroup, outputting the synchronization deviation; amplitude deviation, waveform residual, and correlation decrease are extracted based on the difference between the final waveform and the group reference waveform, which together constitute four types of error state components; a comprehensive health index is obtained by normalization based on the dynamic sub-operating condition baseline, and the weighted continuous anomaly ratio, trend correction slope, and comprehensive degradation score are calculated to realize short-term anomaly alarm and long-term degradation trend assessment; no external standard is required, and it has the advantages of strong robustness, good interpretability, and operating condition adaptability.

[0039] Please see Figure 2 The diagram shows a structural block diagram of an online quantitative analysis system for the error status of an optical fiber current transformer according to this application.

[0040] like Figure 2 As shown, the online quantitative analysis system 200 for the error status of fiber optic current transformers includes an acquisition module 210, a screening module 220, an alignment module 230, an output module 240, an extraction module 250, a normalization module 260, and an analysis module 270.

[0041] The acquisition module 210 is configured to acquire and preprocess the current waveform data of each current transformer, and filter out an effective time window based on an effective value threshold. The filtering module 220 is configured to calculate the weighted coupling difference between every two current transformers within the effective time window, calculate the individual reliability of each current transformer in the current time window based on the weighted coupling difference, and then perform weighted fusion with the historical recursive reliability of the current transformer to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically selected. The alignment module 230 is configured to select the current transformer with the largest recursive reliability from all current transformers as a temporary reference current transformer, and perform a first time-delay alignment of the current waveform of each current transformer relative to the temporary reference current transformer to obtain a coarse alignment waveform. The output module 240 is configured to construct a group reference waveform using the coarse alignment waveforms of each current transformer in the reliable reference subgroup, and output the coarse alignment waveform of each current transformer relative to the group reference waveform. The waveform undergoes a second time-delay alignment to obtain a fine alignment correction. The sum of the time delay from the first time-delay alignment and the fine alignment correction from the second time-delay alignment is taken as the total alignment amount, which is output as the synchronization deviation. Simultaneously, the final aligned waveform is obtained based on the fine alignment correction. The extraction module 250 is configured to extract the amplitude deviation, waveform residual, and correlation decrease value based on the difference between the final aligned waveform and the group reference waveform. These, along with the synchronization deviation, constitute the error state component. The normalization module 260 is configured to normalize the error state component based on the dynamic sub-operating condition baseline to obtain the normalized deviation of each current transformer. The normalized deviation of each current transformer is then weighted and summed to obtain the comprehensive health index of each current transformer. The analysis module 270 is configured to calculate the weighted continuous abnormality ratio, trend correction slope, and comprehensive degradation score based on the historical time window sequence of the comprehensive health index of the current transformer, for short-term abnormality alarm and long-term degradation trend assessment of individual current transformers.

[0042] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0043] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the online quantitative analysis method for the error state of the fiber optic current transformer in any of the above method embodiments. In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows: Acquire the current waveform data of each current transformer and perform preprocessing, then filter out the effective time window based on the effective value threshold; Within the effective time window, the weighted coupling difference between every two current transformers is calculated. Based on the weighted coupling difference, the individual reliability of each current transformer in the current time window is calculated. Then, the individual reliability is weighted and fused with the current transformer's own historical recursive reliability to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically selected. Select the current transformer with the highest recursive confidence from all transformers as the temporary reference transformer, and perform the first time-delay alignment of the current waveforms of each transformer relative to the temporary reference transformer to obtain the coarse alignment waveform. A group reference waveform is constructed using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup. The coarse alignment waveforms of each mutual inductor are then re-aligned with the group reference waveform to obtain a fine alignment correction. The sum of the delay from the first time-delay alignment and the fine alignment correction from the second time-delay alignment is taken as the total alignment amount. The total alignment amount is output as the synchronization deviation. The final alignment waveform is obtained based on the fine alignment correction. Based on the difference between the final aligned waveform and the group reference waveform, the amplitude deviation, waveform residual and correlation decrease value are extracted, and together with the synchronization deviation, they constitute the error state component. Based on the dynamic working condition baseline, the error state components are normalized respectively to obtain the normalized deviation of each current transformer, and the normalized deviation of each current transformer is weighted and summed to obtain the comprehensive health index of each current transformer. Based on the historical time window series of the comprehensive health index of the current transformer, the weighted continuous abnormality ratio, trend correction slope and comprehensive degradation score are calculated to perform short-term abnormality alarm and long-term degradation trend assessment for a single current transformer.

[0044] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the online quantitative analysis system for fiber optic current transformer error status, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely configured relative to a processor, which can be connected to the online quantitative analysis system for fiber optic current transformer error status via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0045] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby realizing the online quantitative analysis method for the error state of the fiber optic current transformer described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the online quantitative analysis system for the error state of the fiber optic current transformer. The output device 340 may include a display screen or other display device.

[0046] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0047] In one implementation, the above-described electronic device is applied in an online quantitative analysis system for the error status of fiber optic current transformers, serving as a client, and includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Acquire the current waveform data of each current transformer and perform preprocessing, then filter out the effective time window based on the effective value threshold; Within the effective time window, the weighted coupling difference between every two current transformers is calculated. Based on the weighted coupling difference, the individual reliability of each current transformer in the current time window is calculated. Then, the individual reliability is weighted and fused with the current transformer's own historical recursive reliability to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically selected. Select the current transformer with the highest recursive confidence from all transformers as the temporary reference transformer, and perform the first time-delay alignment of the current waveforms of each transformer relative to the temporary reference transformer to obtain the coarse alignment waveform. A group reference waveform is constructed using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup. The coarse alignment waveforms of each mutual inductor are then re-aligned with the group reference waveform to obtain a fine alignment correction. The sum of the delay from the first time-delay alignment and the fine alignment correction from the second time-delay alignment is taken as the total alignment amount. The total alignment amount is output as the synchronization deviation. The final alignment waveform is obtained based on the fine alignment correction. Based on the difference between the final aligned waveform and the group reference waveform, the amplitude deviation, waveform residual and correlation decrease value are extracted, and together with the synchronization deviation, they constitute the error state component. Based on the dynamic working condition baseline, the error state components are normalized respectively to obtain the normalized deviation of each current transformer, and the normalized deviation of each current transformer is weighted and summed to obtain the comprehensive health index of each current transformer. Based on the historical time window series of the comprehensive health index of the current transformer, the weighted continuous abnormality ratio, trend correction slope and comprehensive degradation score are calculated to perform short-term abnormality alarm and long-term degradation trend assessment for a single current transformer.

[0048] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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. A method for online quantitative analysis of the error state of an optical fiber current transformer, characterized in that, The method, applicable to redundant current measurement scenarios involving n fiber optic current transformers at the same measurement point, where n ≥ 3, includes: Acquire the current waveform data of each current transformer and perform preprocessing, then filter out the effective time window based on the effective value threshold; Within the effective time window, the weighted coupling difference between every two current transformers is calculated. Based on the weighted coupling difference, the individual reliability of each current transformer in the current time window is calculated. Then, the individual reliability is weighted and fused with the current transformer's own historical recursive reliability to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically selected. Select the current transformer with the highest recursive confidence from all transformers as the temporary reference transformer, and perform the first time-delay alignment of the current waveforms of each transformer relative to the temporary reference transformer to obtain the coarse alignment waveform. A group reference waveform is constructed using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup. The coarse alignment waveforms of each mutual inductor are then re-aligned with the group reference waveform to obtain a fine alignment correction. The sum of the delay from the first time-delay alignment and the fine alignment correction from the second time-delay alignment is taken as the total alignment amount. The total alignment amount is output as the synchronization deviation. The final alignment waveform is obtained based on the fine alignment correction. Based on the difference between the final aligned waveform and the group reference waveform, the amplitude deviation, waveform residual and correlation decrease value are extracted, and together with the synchronization deviation, they constitute the error state component. Based on the dynamic working condition baseline, the error state components are normalized respectively to obtain the normalized deviation of each current transformer, and the normalized deviation of each current transformer is weighted and summed to obtain the comprehensive health index of each current transformer. Based on the historical time window series of the comprehensive health index of the current transformer, the weighted continuous abnormality ratio, trend correction slope and comprehensive degradation score are calculated to perform short-term abnormality alarm and long-term degradation trend assessment for a single current transformer.

2. The online quantitative analysis method for the error state of an optical fiber current transformer according to claim 1, characterized in that, The calculation involves calculating the weighted coupling difference between every two current transformers, calculating the individual reliability of each current transformer within the current time window based on the weighted coupling difference, and then weighting and fusing the individual reliability with the historical recursive reliability of the current transformer itself to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically selected, including: Based on the current amplitude of the i-th current transformer within the p-th effective time window The current amplitude of the j-th current transformer within the p-th time window Waveform correlation coefficient and optimal alignment delay The weighted coupling difference between the i-th and j-th current transformers is calculated using the following expression: ， In the formula, For reference amplitude, Let be the time delay normalization constant. These are weighted coefficients, and their sum is 1. It is a very small positive number; The individual reliability of the i-th current transformer within the p-th effective time window is calculated based on the weighted coupling difference. The expression is: ， In the formula, This represents the total number of current transformers. It is an exponential function with the natural constant e as its base. This represents the summation of all j except i; By using a time memory factor to weight and fuse the current individual confidence level with the historical recursive confidence level of the previous effective time window of the current transformer, the recursive confidence level of the current effective time window is obtained, expressed as: ， In the formula, Let i be the recursive reliability of the i-th current transformer within the p-th valid time window. As a time memory factor, , The historical recursive reliability of the i-th current transformer within the p-1-th effective time window; Calculate the mean recursive reliability of all current transformers within the current valid time window. and standard deviation And determine the trusted reference subgroup according to the preset filtering rules. The expression for the filtering rule is: ， In the formula, This is the adjustment coefficient.

3. The online quantitative analysis method for the error state of an optical fiber current transformer according to claim 1, characterized in that, The step of selecting the current transformer with the highest recursive confidence from all current transformers as a temporary reference current transformer, and performing a first time-delay alignment of the current waveforms of each current transformer relative to the temporary reference current transformer to obtain a coarsely aligned waveform includes: Select the current transformer with the highest recursive confidence level from all current transformers as the temporary reference current transformer. The numbering of the temporary reference current transformer satisfies the following: ， In the formula, Let i be the recursive reliability of the i-th current transformer within the p-th valid time window. This is the number of the temporary reference transformer. In order to make The value of i when the maximum value is reached; Calculate the coarse alignment delay of the i-th current transformer relative to the temporary reference current transformer. The expression for the coarse alignment delay is: ， ， In the formula, The coarse alignment delay of the i-th current transformer relative to the temporary reference current transformer is given by: For time delay variables, After removing DC from the i-th current transformer, Current waveform at time t, For the temporary reference transformer in the p-th effective time window Waveform at time, After removing DC power from the kth current transformer, Current waveform at time t, It is a cross-correlation function; The coarse-aligned waveform is obtained by shifting the coarse-aligned delay waveform, and the expression is: ， In the formula, This is the coarse alignment waveform of the i-th current transformer after the first time-delay alignment. For time variables, Let i be the waveform function after DC removal from the i-th current transformer. This indicates that the original waveform is shifted along the time axis. ,when The waveform shifts to the left when it is positive. When the value is negative, the waveform shifts to the right.

4. The online quantitative analysis method for the error state of an optical fiber current transformer according to claim 1, characterized in that, The process involves constructing a group reference waveform using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup, performing a second time delay alignment on the coarse alignment waveforms of each mutual inductor relative to the group reference waveform to obtain a fine alignment correction, and using the sum of the time delay from the first time delay alignment and the fine alignment correction from the second time delay alignment as the total alignment amount. This total alignment amount is then output as the synchronization deviation. Simultaneously, the final aligned waveform is obtained based on the fine alignment correction amount, including: By trusted reference subgroup The coarse-aligned waveform is used to construct a group reference waveform by taking the median point by point. The expression is: ， In the formula, For group reference waveform, This is the coarse alignment waveform of the i-th current transformer after the first time-delay alignment. To obtain the median, that is, for each fixed time point t, take all... corresponding The median; The fine alignment correction for the i-th current transformer relative to the group reference waveform is calculated using the following expression: ， In the formula, Let be the fine alignment correction amount of the i-th current transformer relative to the group reference waveform. For the group reference waveform in time shift The waveform after that, For time delay variables, It is a cross-correlation function; The total alignment amount is obtained by adding the coarse alignment delay and the fine alignment correction. The total alignment amount is output as the synchronization deviation, and the final alignment waveform is obtained based on the fine alignment correction amount, expressed as: ， In the formula, This represents the final aligned waveform of the i-th current transformer after two steps of coarse and fine alignment correction within the p-th effective time window. For time variables, Let i be the coarse alignment waveform function for the i-th mutual inductor. This indicates that the coarse-aligned waveform will be shifted on the time axis. ,when The waveform shifts to the left when it is positive. When the value is negative, the waveform shifts to the right.

5. The online quantitative analysis method for the error state of an optical fiber current transformer according to claim 4, characterized in that, The expression for calculating the amplitude deviation is: ， ， ， In the formula, For amplitude deviation, This refers to the relative amplitude deviation. The effective value of the group reference waveform, To find the root mean square value; The expression for calculating the waveform residual is as follows: ， In the formula, For waveform residuals; The expression for calculating the correlation decrease value is: ， In the formula, This represents the decrease in correlation. It is a cross-correlation function; The expression for calculating the synchronization deviation is: ， In the formula, This is a synchronization deviation.

6. The online quantitative analysis method for the error state of an optical fiber current transformer according to claim 1, characterized in that, Based on the dynamic working condition baseline, the error state components are normalized to obtain the normalized deviation of each current transformer, including: Establish the operating condition category for the current time window. , This is the working condition mapping function. For ambient temperature, For the load range, The effective value of the group reference waveform; For the j-th type of error state component, the baseline center and baseline scale are updated respectively, as expressed by: ， ， In the formula, , These are the center estimate and scale estimate, respectively, obtained from historical health samples under the current operating condition category. , To update the coefficients, and the values ​​range from 0 to 1, , Let p be the baseline center and baseline scale for the p-th effective time window. , The baseline center and baseline scale for the (p-1)th effective time window; Calculate the normalized deviation of the type j error state component of the i-th current transformer within the p-th time window based on the baseline center and baseline scale. The expression is: ， In the formula, Let be the original quantity of the j-th type of error state component of the i-th current transformer. It is a very small positive number.

7. The online quantitative analysis method for the error state of an optical fiber current transformer according to claim 1, characterized in that, The expression for calculating the weighted proportion of persistent anomalies is: ， In the formula, Let be the weighted continuous anomaly ratio of the i-th current transformer over the most recent N time windows. To calculate the window length, This refers to the time window number. The comprehensive health index of the i-th current transformer in the r-th time window. Forgetting factor, As the warning threshold, It is an indicator function; The expression for calculating the trend correction slope is: ， ， ， In the formula, Let be the trend correction slope of the i-th current transformer over the most recent N time windows. The arithmetic mean of the time sequence numbers. Let be the arithmetic mean of the comprehensive health index of the i-th current transformer over the most recent N time windows. This is the preset fluctuation correction coefficient. The comprehensive health index of the i-th current transformer in the (r-1)-th time window; The expression for calculating the comprehensive degradation score based on the weighted persistent abnormality ratio and the trend correction slope is as follows: ， In the formula, This represents the overall degradation score of the i-th current transformer over the most recent N time windows. Let be the median of the comprehensive health index of the i-th current transformer over the most recent N time windows. , , These are the preset weighting coefficients.

8. An online quantitative analysis system for the error status of an optical fiber current transformer, characterized in that, The system is applicable to redundant current measurement scenarios involving n fiber optic current transformers at the same measurement point, where n ≥ 3. The system includes: The acquisition module is configured to acquire the current waveform data of each current transformer and perform preprocessing, and filter out the effective time window based on the effective value threshold. The filtering module is configured to calculate the weighted coupling difference between every two current transformers within the effective time window, calculate the individual reliability of each current transformer in the current time window based on the weighted coupling difference, and then perform weighted fusion with the current transformer's own historical recursive reliability to obtain the recursive reliability of each current transformer. Based on the mean and standard deviation of the recursive reliability of all current transformers, a reliable reference subgroup is dynamically filtered out. The alignment module is configured to select the one with the highest recursive confidence from all current transformers as a temporary reference current transformer, and perform the first time-delay alignment of the current waveforms of each current transformer relative to the temporary reference current transformer to obtain a coarse alignment waveform. The output module is configured to construct a group reference waveform using the coarse alignment waveforms of each mutual inductor in the trusted reference subgroup, and perform a second time delay alignment on the coarse alignment waveforms of each mutual inductor relative to the group reference waveform to obtain a fine alignment correction amount. The sum of the time delay of the first time delay alignment and the fine alignment correction amount of the second time delay alignment is taken as the total alignment amount. The total alignment amount is output as the synchronization deviation, and the final alignment waveform is obtained based on the fine alignment correction amount. The extraction module is configured to extract amplitude deviation, waveform residual and correlation decrease value based on the difference between the final aligned waveform and the group reference waveform, and together with the synchronization deviation, they constitute the error state component. The normalization module is configured to normalize the error state components based on the dynamic working condition baseline to obtain the normalized deviation of each current transformer, and then weighted summation of the normalized deviation of each current transformer to obtain the comprehensive health index of each current transformer. The analysis module is configured to calculate the weighted continuous abnormality ratio, trend correction slope, and comprehensive degradation score based on the historical time window sequence of the comprehensive health index of the current transformer, so as to perform short-term abnormality alarms and long-term degradation trend assessments for a single current transformer.

9. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1 to 7.

Images