Current transformer error state discrimination method and device, terminal device and medium

Abstract

The application discloses a current transformer error state discrimination method and device, terminal equipment and medium, and the method comprises the following steps: based on each line under the same node of a transformer substation, according to the real-time secondary output data of the measured current transformer in each line and the historical data under the normal operation state, a monitoring data set and a modeling data set are respectively constructed; a line fault state is simulated in the modeling data set to generate fault data; an abnormal database of the measured current transformer is constructed by using the fault data and the modeling data set; a local anomaly factor threshold boundary of the abnormal database is determined; the local anomaly factor of the monitoring data set is calculated and compared with the local anomaly factor threshold boundary to discriminate the error state of the measured current transformer. The application realizes online evaluation of the current transformer measurement error, gets rid of the dependence on power-off and physical standard devices, is applicable to current transformers of different principles or accuracy levels, and has the advantages of high precision and strong applicability.

Description

Technical Field

[0001] This invention relates to the field of power grid data analysis technology, and in particular to a method, device, terminal equipment, and medium for judging the error state of a current transformer. Background Technology

[0002] A current transformer (CT) is a general-purpose device that converts large currents into smaller currents for measurement, and it is widely used in power systems. Metering performance is the most critical attribute of a current transformer. Over long-term operation, current transformers are affected by various physical fields such as electricity, magnetism, heat, and force, which can cause errors or even malfunctions, leading to current measurement failures and seriously affecting the stable operation of the power system. Therefore, current transformers are usually calibrated before being connected to the grid to ensure their measurement accuracy.

[0003] Currently, the verification of current transformers typically falls into two categories: online error assessment based on precise modeling and signal processing. Precise modeling relies on a known, precise equivalent model of the power system to establish a system of equations to solve for unknown transformer errors. However, the high-precision equivalent parameters of the power grid required for this method are often difficult to obtain, making engineering applications challenging. Signal processing methods typically utilize wavelet transform, frequency shift algorithms, and integrated filtering to analyze the output signal of a single transformer and identify error changes. However, this method can only detect abrupt errors in the CT and is not sensitive to gradual changes in errors over long-term operation. It suffers from the problems of impacting the power grid and low measurement accuracy, resulting in limited engineering applicability. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, terminal equipment, and medium for judging the error state of a current transformer, so as to solve the problems of low measurement accuracy, poor feasibility, and easy impact on the power grid in the existing online identification methods for the error state of current transformers.

[0005] To achieve the above objectives, the present invention provides a method for determining the error state of a current transformer, comprising:

[0006] Based on each line under the same node of the substation, a monitoring dataset and a modeling dataset are constructed according to the real-time secondary output data of the current transformer under test in each line and the historical data under normal operation.

[0007] Simulate line fault conditions in the modeling dataset to generate fault data; use the fault data and the modeling dataset to construct an anomaly database for the current transformer under test.

[0008] Determine the local anomaly factor threshold boundaries of the anomaly database;

[0009] Calculate the local anomaly factor of the monitoring dataset and compare it with the threshold boundary of the local anomaly factor to determine the error state of the current transformer under test.

[0010] Furthermore, based on each line under the same node of the substation, a monitoring dataset and a modeling dataset are constructed according to the real-time secondary output data of the current transformer under test and the historical data under normal operation in each line. This includes: collecting the real-time secondary output data of the current transformer under test, extracting the fundamental current amplitude and phase in the real-time secondary output data, constructing the real part and imaginary part of the current vector according to the current amplitude and phase, and using the constructed current vector as the monitoring dataset.

[0011] Historical output data of the current transformer under test under normal operating conditions are collected. The real and imaginary parts of the current vector are constructed based on the current amplitude and phase in the historical output data. The constructed current vector is used as the modeling dataset.

[0012] Furthermore, after constructing the monitoring dataset and the modeling dataset respectively, the method further includes data preprocessing of the monitoring dataset and the modeling dataset, including:

[0013] Using Kirchhoff's current law, the node current vector sums in the monitoring dataset and the modeling dataset are calculated respectively; the node current vector sum is the sum of the current vectors of all lines in the node.

[0014] The line with the largest current vector amplitude among all lines is used as the normalization reference. The sum of the node current vectors is divided by the normalization reference to obtain the normalized current vector of the node.

[0015] Furthermore, the step of simulating line fault states in the modeling dataset to generate fault data includes:

[0016] Based on the modeling dataset, critical errors are added to the modeling data according to preset conditions to generate fault data for each line.

[0017] Further, determining the local anomaly factor threshold boundary of the anomaly database includes:

[0018] Based on the aforementioned anomaly database, multiple local anomaly factors are calculated for each line when a fault occurs, and the smallest local anomaly factor among these multiple local anomaly factors is taken as the local anomaly factor threshold for the corresponding line.

[0019] The minimum value among the local anomaly factor thresholds is used as the local anomaly factor threshold boundary of the anomaly database.

[0020] Further, the calculation of local anomaly factors in the monitoring dataset and comparison with the threshold boundary of the local anomaly factors to determine the error state of the current transformer under test includes:

[0021] When the local anomaly factor of the monitoring dataset is greater than the threshold boundary of the local anomaly factor, it is determined that the current transformer under test has exceeded the error limit and a fault warning is triggered.

[0022] The present invention also provides a current transformer error state discrimination device, comprising:

[0023] The data acquisition unit is used to construct monitoring datasets and modeling datasets based on the real-time secondary output data of the current transformer under test in each line under the same node of the substation and the historical data under normal operation.

[0024] The database construction unit is used to simulate line fault states in the modeling dataset and generate fault data; and to construct an anomaly database of the current transformer under test using the fault data and the modeling dataset.

[0025] A threshold boundary determination unit is used to determine the local anomaly factor threshold boundaries of the anomaly database;

[0026] The error state discrimination unit is used to calculate the local anomaly factor of the monitoring dataset and compare it with the threshold boundary of the local anomaly factor to determine the error state of the current transformer under test.

[0027] Furthermore, the current transformer error state discrimination device further includes: a data preprocessing unit, used for:

[0028] Using Kirchhoff's current law, the node current vector sums in the monitoring dataset and the modeling dataset are calculated respectively; the node current vector sum is the sum of the current vectors of all lines in the node.

[0029] The line with the largest current vector amplitude among all lines is used as the normalization reference. The sum of the node current vectors is divided by the normalization reference to obtain the normalized current vector of the node.

[0030] The present invention also provides a terminal device, comprising:

[0031] One or more processors;

[0032] A memory, coupled to the processor, for storing one or more programs;

[0033] When the one or more programs are executed by the one or more processors, the one or more processors implement the current transformer error state discrimination method as described in any of the preceding claims.

[0034] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the current transformer error state discrimination method as described in any of the preceding claims.

[0035] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0036] This invention discloses a method for judging the error state of a current transformer, comprising: constructing a monitoring dataset and a modeling dataset based on each line under the same node of a substation, according to the real-time secondary output data and historical data under normal operation of the current transformer under test in each line; simulating line fault states in the modeling dataset to generate fault data; constructing an anomaly database of the current transformer under test using the fault data and the modeling dataset; determining the threshold boundary of the local anomaly factor in the anomaly database; calculating the local anomaly factor in the monitoring dataset and comparing it with the threshold boundary of the local anomaly factor to judge the error state of the current transformer under test. This invention realizes online evaluation of the metering error of current transformers, eliminating the dependence on power outages and physical standards, and is applicable to current transformers of different principles or accuracy levels, with advantages such as high accuracy and strong applicability. Attached Figure Description

[0037] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart illustrating a current transformer error state discrimination method provided in a certain embodiment of the present invention;

[0039] Figure 2 This is a flowchart illustrating an online monitoring method for current transformer faults provided in a certain embodiment of the present invention.

[0040] Figure 3 This is a diagram illustrating the effect of local anomaly factor identification according to a certain embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram of the structure of a current transformer error state discrimination device provided in a certain embodiment of the present invention;

[0042] Figure 5 This is a schematic diagram of the structure of a current transformer error state discrimination device provided in another embodiment of the present invention;

[0043] Figure 6 This is a schematic diagram of the structure of a terminal device provided in a certain embodiment of the present invention. Detailed Implementation

[0044] 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, and 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.

[0045] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0046] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0047] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0048] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0049] Please see Figure 1 One embodiment of the present invention provides a method for determining the error state of a current transformer. For example... Figure 1 As shown, the current transformer error state discrimination method includes steps S10 to S40. The specific steps are as follows:

[0050] S10. Based on each line under the same node of the substation, a monitoring dataset and a modeling dataset are constructed respectively according to the real-time secondary output data of the current transformer under test in each line and the historical data under normal operation.

[0051] S20. Simulate line fault conditions in the modeling dataset to generate fault data; use the fault data and the modeling dataset to construct an anomaly database for the current transformer under test.

[0052] S30. Determine the local anomaly factor threshold boundaries of the anomaly database;

[0053] S40. Calculate the local anomaly factor of the monitoring dataset and compare it with the threshold boundary of the local anomaly factor to determine the error state of the current transformer under test.

[0054] It should be noted that current transformers typically undergo mandatory calibration before being put into operation to ensure the accuracy of their measurements. Initially, the calibration method for current transformers was mainly based on "power outage calibration," which involved periodically shutting down power to perform calibration. However, this method was difficult to implement due to the limited number and duration of planned power outages, making it impossible to achieve full coverage of calibrations within a given period, resulting in a large number of overdue calibrations. To break free from the dependence on power outages, the "live-line calibration" method was proposed. Its basic principle is similar to the "power outage calibration" method, using a standard to obtain a relative true value. The deviation between the measured value of the transformer under calibration and the relative true value is the transformer's error. The difference is that the physical standard used in the "live-line calibration" method is specially designed, featuring small size and light weight. It can be connected to the same circuit as the transformer under calibration using a live-line operation method, enabling short-term online operation. Clearly, this method is still limited by the physical standard and can only be performed periodically. Furthermore, live-line operation may trigger local transient processes, posing safety hazards to operators, verification equipment, and even the operation of the power system. Therefore, this embodiment aims to provide a new method to overcome the shortcomings of the two aforementioned verification methods.

[0055] First, the error monitoring method for voltage transformers is explained. Voltage transformer error monitoring primarily relies on cyber-physical fusion. This method takes into account the fact that in actual power grid operation, the majority of transformers are functioning normally, while a very small minority are malfunctioning. It uses a group of transformers with electrical physical connections within the same substation as the evaluation object, and the physical correlation within the group as a constraint. Through multivariate statistical methods, it collaboratively analyzes the information contained in the real-time large-scale data of secondary measurements output by the transformer group, uncovering the real-time status of the tracking constraint relationships. This information fusion approach completely eliminates the need for power outages and physical standards, and does not rely on precise equivalent parameters. Error deviation monitoring can be achieved using only the collected data, making it highly applicable in engineering projects.

[0056] However, this method is only applicable to voltage transformers and not current transformers because: First, current transformers do not have strong coupling relationships, while voltage transformers do; second, the measured values ​​of voltage transformers measuring the same phase at the same node in a substation should be consistent, which can be determined by comparing mutual measurements, while line currents are independent and cannot be compared; third, in a steady-state substation, the voltage amplitude changes by 110%-120% of the rated voltage, with relatively small voltage fluctuations; however, the current changes of each line measured by current transformers in a substation are independent, with amplitudes varying independently from 0%-120% of the rated current, resulting in large fluctuations. Therefore, the method provided in this embodiment, based on the concept of cyber-physical fusion, proposes to take the current transformers of multiple lines at the same characteristic node as the evaluation object, use rigid constraints, including Kirchhoff's current law, as constraints, and cluster analysis of node current vector sums, thereby enabling online monitoring of the error state of the current transformers.

[0057] Specifically, in step S10, the current transformers (CTs) of each line at the same node in the substation are first grouped together, and the real-time secondary output data of the CT group is collected to construct a monitoring dataset. Simultaneously, historical data of the CTs under normal operating conditions is also collected in this step to construct a modeling dataset. It should be noted that, for ease of understanding, this embodiment mainly focuses on a specific node in the substation and the current transformers of all lines on that node. In practical applications, the method provided by this invention is applicable to any node in the substation, and no limitations are imposed here.

[0058] In a specific embodiment, step S10 further includes:

[0059] 1.1) Collect the real-time secondary output data of the current transformer under test, extract the current amplitude and phase of the fundamental wave in the real-time secondary output data, construct the real part and imaginary part of the current vector according to the current amplitude and phase respectively, and use the constructed current vector as the monitoring dataset.

[0060] 1.2) Collect historical output data of the current transformer under normal operation. Construct the real part and imaginary part of the current vector based on the current amplitude and phase in the historical output data. Use the constructed current vector as the modeling dataset.

[0061] In this embodiment, the current vector of the current substation node at each moment is mainly represented by the imaginary part and the real part. Specifically, the real part and the imaginary part of the current vector are constructed by the current amplitude and the phase, respectively. Then, the obtained current vector is used as a two-dimensional modeling dataset and a two-dimensional monitoring dataset.

[0062] In one embodiment, after executing step S10 and before executing step S20, the method further includes data preprocessing of the monitoring dataset and the modeling dataset, including:

[0063] Using Kirchhoff's current law, the node current vector sums in the monitoring dataset and the modeling dataset are calculated respectively; the node current vector sum is the sum of the current vectors of all lines in the node.

[0064] The line with the largest current vector amplitude among all lines is used as the normalization reference. The sum of the node current vectors is divided by the normalization reference to obtain the normalized current vector of the node.

[0065] In this embodiment, the current vector sum of each line node is obtained using Kirchhoff's Current Law (KCL). Then, the line with the largest current vector amplitude among all lines under that node is selected, and its current vector is used as the normalization reference. Preferably, the current vector of the transformer branch is used as the normalization standard. Then, the obtained node current vector sum is divided by the normalization reference to obtain the normalized current vector of the node at that moment.

[0066] Further, step S20 is performed to construct an anomaly database for the current transformer under test. Specifically, in step S20, parameter selection is first performed, and an appropriate neighborhood parameter K is selected based on the sampling frequency and evaluation period of the CT secondary side data. Then, the single-line fault situation is simulated on the modeling dataset to generate fault data for each line. Finally, the fault data for each line and the two-dimensional modeling dataset are combined to construct an anomaly database of single-line CT critical deviation.

[0067] In one specific embodiment, the generation of fault data for each line in step S20 mainly involves: based on the modeling dataset, adding critical errors to the modeling data according to preset conditions to generate fault data for each line. It should be noted that these preset conditions primarily determine the accuracy class of the current transformer, then divide this accuracy class by a certain value to obtain the measurement error variation limit of the current transformer, and finally add critical errors to each modeling dataset according to this measurement error variation limit to generate fault data for each line.

[0068] Further, step S30 is performed, namely, determining the threshold boundary of the local anomaly factor (hereinafter referred to as LOF factor) of the anomaly database.

[0069] In a specific embodiment, step S30 specifically includes:

[0070] 3.1) Based on the aforementioned anomaly database, calculate the LOF factor for each line fault according to line classification, and use the minimum LOF factor of each faulty line as the LOF factor threshold for the corresponding line.

[0071] 3.2) The minimum value among all LOF factor thresholds is used as the LOF factor threshold boundary for the abnormal database.

[0072] It should be explained that in this step, based on the critical out-of-tolerance anomaly database of single-line current transformers, LOF analysis is performed on the critical out-of-tolerance anomaly database according to line classification. The minimum LOF factor obtained when each line is faulty is used as the out-of-group threshold for that line, and the out-of-group thresholds for critical out-of-tolerance anomalies of each line are obtained. The minimum value of the thresholds is taken as the local out-of-group factor threshold boundary.

[0073] Finally, step S40 is executed, where the two-dimensional monitoring dataset is processed using the local outlier algorithm to calculate the LOF factor of the monitoring dataset. This LOF factor is then compared with the LOF factor threshold boundary to determine the error state of the current transformer under test. Specifically, if the LOF factor of the monitoring dataset is greater than the LOF factor threshold boundary, it is determined that the current transformer under test has exceeded the error limit, and a fault warning is triggered.

[0074] Therefore, the current transformer error state discrimination method provided in this embodiment of the invention realizes online evaluation of current transformer measurement error, gets rid of dependence on power outages and physical standards, and can be applied to current transformers of different principles or accuracy levels. It has the advantages of high accuracy and strong applicability.

[0075] To aid in understanding the method provided by this invention, in a specific embodiment, a current transformer is used as the object under test to illustrate the fault diagnosis method provided by this invention. Figure 2 This is a flowchart of the online monitoring method for current transformer faults provided in this embodiment.

[0076] like Figure 2 As shown, the circuit model in this embodiment is a circuit feature node model with one input line and five output lines. Specifically, this embodiment includes the following steps:

[0077] (1) Data Acquisition: The current transformers (CTs) of 6 lines at the same node of the substation are grouped together. The current signals on the secondary side of the current transformers of each of the 6 lines at the node are collected in real time through a data acquisition device, including the amplitude and phase of the current signals, thereby forming the current vectors of the 6 lines at the same characteristic node and constructing a monitoring dataset. The normal historical data of the 6 lines are collected to construct a modeling dataset. The data size is as follows: the elements in the modeling dataset are the historical normal current vectors of each line.

[0078] (2) Data Preprocessing: The current vectors on the six lines in the obtained feature node group vector measurement dataset are used to obtain the feature node current data vector sum through the rigid constraint condition: Kirchhoff's Current Law (KCL). The vector sum is then normalized to obtain its normalized vector. In this example, the incoming line is a transformer branch, so the current vector of the incoming line is selected as the normalization standard. The obtained node current vector sum is divided by this normalization standard to obtain the normalized vector of the feature node at this moment. The normalized vector is represented in the form of imaginary and real parts, thus forming a two-dimensional dataset of normalized vectors representing this feature node at different moments. The modeling dataset and the monitoring dataset are both processed according to the above data preprocessing method to obtain their respective two-dimensional modeling dataset and two-dimensional monitoring dataset.

[0079] (3) Parameter Selection and Database Construction: For the measurement error status assessment of CT, in this example, the sampling period of the data acquisition device is 10 minutes, and the data length corresponding to the assessment period of 8 hours is 48. At this time, K can be set to 48, which is much larger than the duration of common data fluctuations, and can eliminate some of the impact of primary power grid fluctuations. Simulate single-line faults on the two-dimensional modeling dataset, that is, add critical errors to the modeling data according to preset conditions to generate fault data for each line. Specifically, the preset conditions mainly determine the accuracy level of the current transformer, and then obtain the measurement error variation limit of the current transformer by dividing the accuracy level by a certain value. Finally, add critical errors to each modeling data according to the measurement error variation limit to generate fault data for each line. For example, in this embodiment, if the accuracy level of CT is S, then the measurement error variation limit of CT is S / 100. That is, according to the number of lines on the same node and the accuracy level of the current transformer corresponding to each line, the corresponding critical error of the current transformer of each line is added to the data in the two-dimensional modeling dataset corresponding to each line in turn to construct the fault data of each line. The method for adding error is as follows: Based on the accuracy class S of the current transformer for this line, the data Q in the corresponding two-dimensional modeling dataset for this line is calculated as: E = Q * (1 + S / 100) or E = Q * (1 - S / 100) to obtain the fault data E for this line. Since the current transformer has an accuracy class of 0.2, an error of 0.2% is added to the two-dimensional modeling dataset. Fault data for each line are then constructed, and a single-line CT critical exceedance anomaly database is built with the two-dimensional modeling dataset. The database is then categorized by line.

[0080] (4) Threshold boundary determination: Based on the single-line CT critical outlier database, LOF analysis is performed on the critical outlier database according to the line classification. The minimum LOF factor obtained when each line is faulty is taken as the outlier threshold of this line. The outlier thresholds when each line is critically outlier are obtained. The minimum value of the threshold is taken as the local outlier factor threshold boundary, as shown in Table 1.

[0081] Table 1 Offline thresholds for each line

[0082] Critical out-of-tolerance circuit Line 1 Line 2 Line 3 Line 4 Line 5 Line 6 Offline threshold 1.782 1.765 1.768 1.750 1.774 1.778

[0083] As shown in Table 1, the minimum outlier threshold among the 6 lines is 1.75. Therefore, 1.75 is selected as the local outlier factor boundary, which is also the LOF factor threshold boundary.

[0084] (5) Algorithm Processing: The Local Outlier Factor (LOF) algorithm is used to analyze and process the two-dimensional monitoring dataset to obtain the corresponding LOF factor. If an LOF factor exceeding the local outlier factor boundary threshold appears in the dataset, it is determined that there is a current transformer with an excessive change in measurement error among the current transformers on the six lines of this feature node, and a fault alarm is triggered. Specifically, when a current transformer with an excessive change in measurement error appears, some outliers will appear, achieving the following effect: Figure 3 As shown. By Figure 3 It can be seen that the sparsely distributed points (square points) are outliers, demonstrating the excellent processing and analysis effect of this method for gradual errors. When an outlier appears, it is considered that one of the current transformers on the six lines of this characteristic node has exceeded the limit of measurement error, thus triggering a fault alarm.

[0085] Therefore, this embodiment can monitor the metering error of the current transformer online and trigger a fault warning when the monitoring result is a fault, so as to make timely manual intervention, which is conducive to maintaining the stability of the power grid system and reducing the occurrence of accidents.

[0086] Please see Figure 4 An embodiment of the present invention also provides a current transformer error state discrimination device, comprising:

[0087] Data acquisition unit 01 is used to construct monitoring datasets and modeling datasets for each line node based on the real-time secondary output data of the current transformer under test in each line and the historical data under normal operation.

[0088] Database construction unit 02 is used to simulate line fault states in the modeling dataset and generate fault data; and to construct an anomaly database of the current transformer under test using the fault data and the modeling dataset.

[0089] Threshold boundary determination unit 03 is used to determine the local anomaly factor threshold boundary of the anomaly database;

[0090] Error state discrimination unit 04 is used to calculate the local anomaly factor of the monitoring dataset and compare it with the threshold boundary of the local anomaly factor to determine the error state of the current transformer under test.

[0091] In one specific embodiment, the current transformer error state discrimination device further includes a data preprocessing unit 05, such as... Figure 5 As shown. Specifically, the data preprocessing unit 05 is used for:

[0092] Using Kirchhoff's current law, calculate the current vector sum of each line node in the monitoring dataset and the modeling dataset, respectively;

[0093] The monitoring dataset and the modeling dataset are normalized by using the line with the largest current vector amplitude among all lines as the normalization benchmark.

[0094] It is understood that the current transformer error state discrimination device provided in this embodiment of the invention is used to execute the current transformer error state discrimination method as described in any of the above embodiments. This embodiment realizes online evaluation of current transformer metering error, eliminating the dependence on power outages and physical standards, and is applicable to current transformers of different principles or accuracy levels, with the advantages of high accuracy and strong applicability.

[0095] Please see Figure 6 An embodiment of the present invention provides a terminal device, comprising:

[0096] One or more processors;

[0097] A memory, coupled to the processor, for storing one or more programs;

[0098] When the one or more programs are executed by the one or more processors, the one or more processors implement the current transformer error state discrimination method as described above.

[0099] The processor controls the overall operation of the terminal device to complete all or part of the steps of the current transformer error state discrimination method described above. The memory stores various types of data to support the operation of the terminal device. This data may include, for example, instructions for any application or method used to operate on the terminal device, as well as application-related data. The memory can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0100] In an exemplary embodiment, the terminal device may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the current transformer error state discrimination method as described in any of the above embodiments and achieve the same technical effect as the above method.

[0101] In another exemplary embodiment, a computer-readable storage medium including a computer program is also provided. When executed by a processor, the computer program implements the steps of the current transformer error state determination method as described in any of the above embodiments. For example, the computer-readable storage medium may be the aforementioned memory including the computer program, which may be executed by a processor of a terminal device to complete the current transformer error state determination method as described in any of the above embodiments and achieve the same technical effects as the aforementioned method.

[0102] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for determining the error state of a current transformer, characterized in that, include: Based on each line under the same node of the substation, a monitoring dataset and a modeling dataset are constructed according to the real-time secondary output data of the current transformer under test in each line and the historical data under normal operation. Simulate line fault states in the modeling dataset to generate fault data; Using the fault data and the modeling dataset, an anomaly database of the current transformer under test is constructed; Determine the local anomaly factor threshold boundaries of the anomaly database; Calculate the local anomaly factor of the monitoring dataset and compare it with the threshold boundary of the local anomaly factor to determine the error state of the current transformer under test; The process includes, after constructing the monitoring dataset and the modeling dataset respectively, data preprocessing of the monitoring dataset and the modeling dataset, including: Using Kirchhoff's current law, the node current vector sums in the monitoring dataset and the modeling dataset are calculated respectively; the node current vector sum is the sum of the current vectors of all lines in the node. The line with the largest current vector amplitude among all lines is taken as the normalization reference. The sum of the node current vectors is divided by the normalization reference to obtain the normalized current vector of the node. Determining the local anomaly factor threshold boundary of the anomaly database includes: Based on the aforementioned anomaly database, multiple local anomaly factors are calculated for each line when a fault occurs, and the smallest local anomaly factor among these multiple local anomaly factors is taken as the local anomaly factor threshold for the corresponding line. The minimum value among the local anomaly factor thresholds is used as the local anomaly factor threshold boundary of the anomaly database.

2. The current transformer error state discrimination method according to claim 1, characterized in that, The method is based on each line under the same node of the substation. According to the real-time secondary output data of the current transformer under test in each line and the historical data under normal operation, a monitoring dataset and a modeling dataset are constructed respectively. The method includes: collecting the real-time secondary output data of the current transformer under test, extracting the fundamental current amplitude and phase in the real-time secondary output data, constructing the real part and imaginary part of the current vector according to the current amplitude and phase respectively, and using the constructed current vector as the monitoring dataset. Historical output data of the current transformer under test under normal operating conditions are collected. The real and imaginary parts of the current vector are constructed based on the current amplitude and phase in the historical output data. The constructed current vector is used as the modeling dataset.

3. The method for judging the error state of a current transformer according to claim 1, characterized in that, The process of simulating line fault states in the modeling dataset to generate fault data includes: Based on the modeling dataset, critical errors are added to the modeling data according to preset conditions to generate fault data for each line.

4. The method for judging the error state of a current transformer according to claim 1, characterized in that, The calculation of local anomaly factors in the monitoring dataset, and comparison with the threshold boundary of the local anomaly factors, to determine the error state of the current transformer under test, includes: When the local anomaly factor of the monitoring dataset is greater than the threshold boundary of the local anomaly factor, it is determined that the current transformer under test has exceeded the error limit and a fault warning is triggered.

5. A current transformer error state discrimination device, characterized in that, include: The data acquisition unit is used to construct monitoring datasets and modeling datasets based on the real-time secondary output data of the current transformer under test in each line under the same node of the substation and the historical data under normal operation. A database construction unit is used to simulate line fault states in the modeling dataset and generate fault data. Using the fault data and the modeling dataset, an anomaly database of the current transformer under test is constructed; A threshold boundary determination unit is used to determine the local anomaly factor threshold boundary of the anomaly database; wherein, determining the local anomaly factor threshold boundary of the anomaly database includes: calculating multiple local anomaly factors when each line is faulty based on the anomaly database, taking the smallest local anomaly factor among the multiple local anomaly factors as the local anomaly factor threshold of the corresponding line; and taking the minimum value among the local anomaly factor thresholds as the local anomaly factor threshold boundary of the anomaly database. The error state discrimination unit is used to calculate the local anomaly factor of the monitoring dataset and compare it with the threshold boundary of the local anomaly factor to determine the error state of the current transformer under test. The data preprocessing unit is used to calculate the node current vector sum in the monitoring dataset and the modeling dataset respectively using Kirchhoff's current law; the node current vector sum is the sum of the current vectors of all lines in the node; the line with the largest current vector amplitude among all lines is used as the normalization reference, and the node current vector sum is divided by the normalization reference to obtain the normalized current vector of the node.

6. A terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the current transformer error state discrimination method as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the current transformer error state discrimination method as described in any one of claims 1-4.

Images