Rapid measurement error test system of voltage transformer low-voltage test equipment based on artificial intelligence

Abstract

The invention discloses a voltage transformer low-voltage test equipment metering error rapid test system based on artificial intelligence, which belongs to the technical field of error test of voltage transformers and comprises an interference analysis module, an error analysis module, an acquisition module and a test module. The interference analysis module is used for carrying out anti-environmental electromagnetic interference analysis and determining an anti-interference curve; the error analysis module is used for performing error analysis on the voltage transformer and establishing a reference error test model and an additional error correction model; the acquisition module is used for acquiring corresponding reference acquisition data and additional acquisition data in real time when the voltage transformer is tested; the test module is used for testing the voltage transformer to obtain an anti-interference curve, and adjusting the reference acquisition data and the additional acquisition data according to the anti-interference curve to obtain first acquisition data and second acquisition data; and performing analysis through the reference error test model and the additional error correction model to obtain a reference metering error.

Description

Technical Field

[0001] The invention belongs to the technical field of error testing of voltage transformers, and in particular is a rapid measurement error testing system for low-voltage testing equipment of voltage transformers based on artificial intelligence. Background Art

[0002] Voltage transformer is an important device for measuring voltage in power system, and its measurement accuracy is crucial for the stable operation of power system and electric energy measurement. However, in practical applications, the measurement error of voltage transformer under low voltage conditions has always been a concern.

[0003] In the prior art, the test of the measurement error of the voltage transformer low-voltage test equipment usually relies on the traditional manual test method. This method has problems such as low test efficiency, low test accuracy, and cumbersome data processing. Specifically, the manual test method requires a lot of time for data collection and processing, and is easily affected by human factors, resulting in inaccurate test results. In addition, the traditional test method is difficult to effectively evaluate complex problems such as the nonlinear error of the voltage transformer under low voltage conditions.

[0004] Based on this, the present invention provides a rapid measurement error testing system for voltage transformer low-voltage testing equipment based on artificial intelligence. Summary of the invention

[0005] In order to solve the problems existing in the above-mentioned scheme, the present invention provides a rapid measurement error testing system for voltage transformer low-voltage testing equipment based on artificial intelligence.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The AI-based rapid measurement error test system for voltage transformer low-voltage test equipment includes interference analysis module, error analysis module, acquisition module, and test module;

[0008] The interference analysis module is used to perform anti-environmental electromagnetic interference analysis and determine an anti-interference curve, wherein the horizontal axis of the anti-interference curve is time and the vertical axis is the anti-interference value; the anti-interference value includes 1 and 0; an anti-interference value of 1 indicates that the anti-electromagnetic interference requirements are met; an anti-interference value of 0 indicates that the anti-electromagnetic interference requirements are not met.

[0009] Furthermore, the method for performing environmental electromagnetic interference resistance analysis includes:

[0010] Connecting to an anti-electromagnetic interference device, and obtaining anti-interference data of the anti-electromagnetic interference device in real time when the anti-electromagnetic interference device is working;

[0011] Analyzing the anti-interference data to obtain an anti-interference value of the anti-interference data;

[0012] The time corresponding to the anti-interference value is identified, and an anti-interference curve is generated according to the anti-interference value at each time, wherein the horizontal axis of the anti-interference curve is the time, and the vertical axis is the anti-interference value.

[0013] Furthermore, an anti-electromagnetic unit is provided in the interference analysis module, and the anti-electromagnetic unit is used to resist environmental electromagnetic interference. The anti-electromagnetic unit is used to replace the anti-electromagnetic interference device, and the anti-electromagnetic unit is used to resist environmental electromagnetic interference during testing, and the anti-electromagnetic interference device is no longer connected.

[0014] Furthermore, the method for analyzing the anti-interference data includes:

[0015] An anti-interference analysis model is established, and the expression of the anti-interference analysis model is:

[0016]

[0017] Where: (St, Y) is the input data, St represents the anti-interference data at the corresponding time, t is the subscript, indicating time; Y is the anti-interference requirement; St→Y means that the anti-interference data at the corresponding time meets the anti-interference requirement; the output data is the anti-interference value KG(St, Y), and the anti-interference value is 1 or 0;

[0018] The anti-interference data and the anti-interference requirements are input into the anti-interference analysis model for analysis to obtain the anti-interference value corresponding to the anti-interference data.

[0019] The error analysis module is used to perform error analysis on the voltage transformer to obtain benchmark error analysis data and additional error analysis data, wherein the benchmark error analysis data includes low voltage error analysis data and load error analysis data; a benchmark error test model and an additional error correction model are established based on the benchmark error analysis data and the additional error analysis data respectively; the benchmark error test model is used to analyze the benchmark acquisition data to obtain the metering test error; the additional error correction model is used to correct the metering test error according to the additional acquisition data to obtain the benchmark metering error.

[0020] The acquisition module is used to acquire corresponding reference acquisition data and additional acquisition data in real time when testing the voltage transformer; and send the reference acquisition data and additional acquisition data to the test module.

[0021] The test module is used to test the voltage transformer, obtain an anti-interference curve, adjust the received reference acquisition data and additional acquisition data according to the anti-interference curve, and mark the adjusted reference acquisition data and additional acquisition data as first acquisition data and second acquisition data respectively;

[0022] The first collected data is analyzed by a benchmark error test model to determine the metrology test error; the second collected data and the metrology test error are analyzed by an additional error correction model to obtain a benchmark metrology error, which is then displayed to a user.

[0023] Furthermore, the method for adjusting the received baseline collected data and the additional collected data according to the anti-interference curve includes:

[0024] Identify the collection time corresponding to the baseline collection data and the additional collection data, and generate collection time axes of the baseline collection data and the additional collection data, respectively, marked as the baseline time axis and the additional time axis;

[0025] The time period corresponding to the anti-interference value of 0 in the anti-interference curve is marked as an abnormal time period, and the abnormal time period is marked in the baseline time axis and the additional time axis. The baseline acquisition data and the additional acquisition data corresponding to the abnormal time period in the baseline time axis and the additional time axis are eliminated, and the baseline acquisition data and the additional acquisition data after the data is eliminated are adjusted for integrity and consistency to obtain the adjusted baseline acquisition data and the additional acquisition data.

[0026] Furthermore, the method for adjusting the integrity and consistency of the baseline collected data and the additional collected data after the data is eliminated includes:

[0027] Mark the baseline collected data and the additional collected data after the data is eliminated as baseline eliminated data and additional eliminated data respectively; perform integrity analysis on the baseline eliminated data and the additional eliminated data respectively to obtain the baseline complete data and the additional complete data corresponding to the baseline eliminated data and the additional eliminated data respectively;

[0028] Identify the acquisition time axes corresponding to the baseline complete data and the additional complete data, respectively, and mark them as the baseline complete axis and the additional complete axis; determine the consistent time period based on the baseline complete axis and the additional complete axis;

[0029] The baseline complete data and the additional complete data are adjusted for consistency according to the consistent time period to obtain the adjusted baseline collection data and the additional collection data.

[0030] Furthermore, the method of performing integrity analysis on the benchmark rejection data includes:

[0031] Step SA1: identifying the endpoint data of the benchmark rejection data according to the abnormal time period and marking it as the benchmark endpoint data;

[0032] Step SA2: Determine whether the reference endpoint data meets the integrity requirement and obtain the integrity determination result;

[0033] Step SA3: when the integrity judgment result satisfies the completeness requirement, no integrity adjustment is performed on the reference endpoint data, and the mark of the reference endpoint data is cancelled;

[0034] When the integrity judgment result is that the completeness requirement is not met, the reference endpoint data is discarded, new reference endpoint data is formed, and the process returns to step SA2;

[0035] Step SA4: looping steps SA1 to SA3 until there is no benchmark endpoint data, marking the current benchmark-removed data as benchmark complete data.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention realizes the rapid and accurate testing of the measurement error of the voltage transformer low-voltage test equipment by introducing artificial intelligence technology. Compared with the traditional manual testing method, the present invention can greatly reduce the testing time and improve the testing efficiency. At the same time, since the artificial intelligence algorithm has powerful data processing and analysis capabilities, it can more accurately evaluate the measurement error of the voltage transformer, providing a strong guarantee for the stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] like Figure 1 As shown, the voltage transformer low-voltage test equipment measurement error rapid test system based on artificial intelligence includes an interference analysis module, an error analysis module, an acquisition module and a test module;

[0042] The interference analysis module is used to perform anti-environmental electromagnetic interference analysis and determine an anti-interference curve. The horizontal axis of the anti-interference curve is time, and the vertical axis is the anti-interference value. The anti-interference value includes 1 and 0. The anti-interference value is 1, indicating that the anti-electromagnetic interference requirements are met; the anti-interference value is 0, indicating that the anti-electromagnetic interference requirements are not met.

[0043] In one embodiment, an environmental electromagnetic interference analysis is performed, and environmental electromagnetic interference is resisted during testing based on the existing electromagnetic interference resistance technology, such as configuring corresponding electromagnetic interference equipment, and performing environmental electromagnetic interference resistance through the electromagnetic interference resistance equipment; the configured electromagnetic interference resistance equipment can be connected to the interference analysis module to achieve real-time acquisition of electromagnetic interference data of the electromagnetic interference resistance equipment, and the electromagnetic interference resistance data mainly includes relevant data detected to indicate whether the electromagnetic interference resistance effect meets the test requirements; the obtained electromagnetic interference resistance data is analyzed in real time to determine whether the electromagnetic interference resistance at each time meets the requirements, and the existing technology can be used for direct judgment;

[0044] When the anti-electromagnetic interference requirements are met, set the anti-interference value to 1;

[0045] When the anti-electromagnetic interference requirements are not met, set the anti-interference value to 0;

[0046] An anti-interference curve is generated according to the anti-interference values ​​corresponding to each time, wherein the horizontal axis of the anti-interference curve is time, and the vertical axis is the anti-interference value.

[0047] In one embodiment, when the anti-electromagnetic interference technology meets the requirements, an anti-electromagnetic unit is set, that is, the anti-electromagnetic unit is built into the test system; the anti-electromagnetic unit is used to resist environmental electromagnetic interference; the anti-interference data of the anti-electromagnetic unit is obtained in real time, and an anti-interference curve is generated according to the anti-interference data.

[0048] In one embodiment, a method for analyzing interference-resistant data includes:

[0049] An anti-interference analysis model is established. The anti-interference analysis model is used to analyze whether the anti-interference data meets the anti-interference requirements. The expression of the anti-interference analysis model is: Where: (St, Y) is the input data, St represents the anti-interference data at the corresponding time, t is the subscript, indicating time; Y is the anti-interference requirement; St→Y means that the anti-interference data at the corresponding time meets the anti-interference requirement, which is judged based on the existing numerical, graphical and other recognition and judgment technologies. It is a built-in analysis and is not reflected in the above model; the output data is the anti-interference value KG(St, Y);

[0050] The anti-interference data and anti-interference requirements are input into the anti-interference analysis model for analysis to obtain the anti-interference value corresponding to the corresponding anti-interference data.

[0051] The error analysis module is used to perform error analysis on the voltage transformer to obtain benchmark error analysis data and additional error analysis data, wherein the benchmark error analysis data includes low voltage error analysis data and load error analysis data; wherein the additional error analysis data refers to error characteristic data caused by environmental factors such as temperature and humidity; the benchmark analysis data refers to error characteristic data of the voltage transformer under low voltage and error characteristic data under different loads;

[0052] A benchmark error test model and an additional error correction model are established based on the benchmark error analysis data and the additional error analysis data respectively; the benchmark error test model is used to subsequently determine the metering error according to the actual low voltage and load conditions of the voltage transformer, and the additional error correction model is used to correct the determined metering error according to temperature, humidity and other data, eliminate the influence of errors such as temperature and humidity, determine the true metering error of the voltage transformer, and mark it as the benchmark metering error; the collected data corresponding to the benchmark error test model is marked as benchmark collected data; the temperature, humidity and other collected data are marked as additional collected data.

[0053] Among them, the baseline error test model and the additional error correction model are both established based on the existing artificial intelligence technology. The training set is set through the corresponding baseline error analysis data and the additional error analysis data, and then the corresponding baseline error test model and the additional error correction model are established; for the additional error correction model, there is currently an intelligent model that can eliminate the influence of corresponding temperature and humidity on the error, and the existing intelligent model can be applied as the additional error correction model.

[0054] Subsequently, a test module is set according to the baseline error test model and the additional error correction model, that is, the test module performs error testing based on the baseline error test model and the additional error correction model.

[0055] For example, for the benchmark error test model: clean the collected benchmark error analysis data to remove invalid or abnormal data. Standardize the benchmark error analysis data so that data of different dimensions can be compared and analyzed. Extract characteristic variables, such as input voltage, output voltage, load size, etc., as input of the intelligent model.

[0056] According to the complexity of the problem and the characteristics of the data, select the appropriate intelligent model. Commonly used intelligent models include neural networks, support vector machines, decision trees, etc. Since there is a complex nonlinear relationship between the error and multiple factors such as input voltage and load, it is best to choose a model with strong nonlinear approximation capabilities such as neural networks.

[0057] The preprocessed benchmark error analysis data is divided into a training set and a test set. The intelligent model is trained using the training set data, and the intelligent model parameters are adjusted to minimize the test error. During the training process, methods such as cross-validation can be used to evaluate the performance of the intelligent model to prevent overfitting or underfitting.

[0058] Use the test set data to verify the trained intelligent model and evaluate the test accuracy and generalization ability of the intelligent model. Perform optimization based on the verification results, such as adjusting the intelligent model structure and adding feature variables. Repeat the training and verification process until the intelligent model performance meets the requirements; mark the intelligent model as the benchmark error test model.

[0059] In one embodiment, the method for obtaining the benchmark error analysis data can be determined according to the prior art.

[0060] In one embodiment, a method for obtaining benchmark error analysis data includes:

[0061] The structural principle data of the voltage transformer is obtained, and the structural principle data includes relevant data such as winding parameters, core material, insulation grade, etc.; the voltage transformer is tested multiple times under low voltage conditions, and the output voltage under different input voltages and the corresponding load data are recorded; the corresponding data processing is performed to ensure the accuracy and completeness of the test data and form benchmark error analysis data; for the load error analysis data, the voltage transformer load error extrapolation model is studied mainly through the excitation equivalent circuit of the voltage transformer, and the error change value of the tested transformer under rated load and lower limit load is calculated, and finally the error characteristics of the voltage transformer are derived.

[0062] The acquisition module is used to acquire corresponding reference acquisition data and additional acquisition data in real time when testing the voltage transformer; and send the reference acquisition data and additional acquisition data to the test module.

[0063] The test module is used to test the voltage transformer, obtain an anti-interference curve, adjust the received reference acquisition data and additional acquisition data according to the anti-interference curve, and mark the adjusted reference acquisition data and additional acquisition data as first acquisition data and second acquisition data respectively;

[0064] The first collected data is analyzed through the benchmark error test model to determine the measurement test error; the second collected data and the measurement test error are analyzed through the additional error correction model to obtain the benchmark measurement error, and the obtained benchmark measurement error is displayed to the user.

[0065] In one embodiment, additional influences may not be considered, that is, additional data is collected, additional error analysis models are not built, and corresponding metrology test errors are directly displayed.

[0066] In one embodiment, the method for adjusting the received baseline acquisition data and the additional acquisition data according to the anti-interference curve includes:

[0067] Identify the collection time corresponding to the baseline collection data and the additional collection data, and generate collection time axes of the baseline collection data and the additional collection data, respectively, marked as the baseline time axis and the additional time axis;

[0068] The time period corresponding to the anti-interference value of 0 in the anti-interference curve is marked as an abnormal time period, and the abnormal time period is marked in the baseline time axis and the additional time axis. The baseline acquisition data and the additional acquisition data corresponding to the abnormal time period in the baseline time axis and the additional time axis are eliminated, and the baseline acquisition data and the additional acquisition data after the data is eliminated are adjusted for integrity and consistency to obtain the adjusted baseline acquisition data and the additional acquisition data.

[0069] In one embodiment, the baseline collected data and the additional collected data after data elimination are adjusted for integrity and consistency, and the adjustment can be performed based on existing technologies.

[0070] In one embodiment, the method for adjusting the integrity and consistency of the baseline collected data and the additional collected data after data removal includes:

[0071] Mark the baseline collected data and the additional collected data after the data is eliminated as baseline eliminated data and additional eliminated data respectively; perform integrity analysis on the baseline eliminated data and the additional eliminated data respectively to obtain the baseline complete data and the additional complete data corresponding to the baseline eliminated data and the additional eliminated data respectively;

[0072] Identify the baseline time axis and the additional time axis corresponding to the baseline complete data and the additional complete data, respectively, and mark them as the baseline complete axis and the additional complete axis for distinction; compare the time periods corresponding to the baseline complete axis and the additional complete axis, respectively, and take the intersection of their time periods as the consistent time period; if they correspond to [1, 8] and [2, 9] respectively, then the consistent time period is [2, 8];

[0073] The baseline complete data and the additional complete data are adjusted for consistency according to the consistent time period, that is, the data outside the consistent time period are eliminated to obtain the adjusted baseline collection data and the additional collection data.

[0074] In one embodiment, a method for performing integrity analysis on baseline culling data includes:

[0075] Step SA1: Identify the endpoint data of the benchmark elimination data according to the abnormal time period and mark it as the benchmark endpoint data; the endpoint data is the data corresponding to the remaining endpoints after eliminating the data corresponding to the abnormal time period. For example, if the abnormal time period is [2, 5], the data corresponding to the time period [2, 5] is eliminated, and an endpoint is generated at the position 2 and 5 respectively. The position data is the endpoint data, such as a knotted rope cut off a section, will produce two endpoints;

[0076] Step SA2: Determine whether the benchmark endpoint data meets the integrity requirements and whether the benchmark endpoint data can meet the error analysis requirements. For example, if the analysis requires the overall data corresponding to the [1.5, 2.1] time period to be analyzed, because [2, 5] is excluded, the endpoint data is incomplete and cannot be used as a whole for error analysis, that is, it does not meet the integrity requirements; that is, whether the integrity requirements are met is determined according to the analysis requirements of the benchmark error test model; and the integrity judgment result is obtained;

[0077] Step SA3: When the integrity judgment result satisfies the completeness requirement, the integrity adjustment of the reference endpoint data is not performed, and the reference endpoint data mark is cancelled;

[0078] When the integrity judgment result is that the completeness requirement is not met, the reference endpoint data is discarded, new reference endpoint data is formed, and the process returns to step SA2;

[0079] Step SA4: looping steps SA1 to SA3 until there is no benchmark endpoint data, marking the current benchmark-removed data as benchmark complete data.

[0080] In one embodiment, whether the reference endpoint data meets the integrity requirement may be determined based on existing technologies.

[0081] In one embodiment, the method for determining whether the reference endpoint data meets the integrity requirement includes:

[0082] Set integrity requirements according to the benchmark error test model, and establish an integrity judgment model based on the integrity requirements to determine whether the corresponding endpoint data meets the integrity requirements;

[0083] The benchmark endpoint data is analyzed through the integrity judgment model to determine whether the benchmark endpoint data meets the integrity requirements.

[0084] In one embodiment, the integrity judgment model can be established based on existing intelligent technology, such as based on a neural network such as a CNN network or a DNN network, and a corresponding training set is established manually for training. The training set includes input data and output data. The input data is the benchmark exclusion data for marking the corresponding benchmark endpoint data, that is, the data adjacent to the benchmark endpoint data needs to be associated, and the output data is the judgment result of whether the benchmark endpoint data meets the integrity requirements; and analysis is performed through the integrity judgment model after successful training.

[0085] In one embodiment, the integrity judgment model is

[0086] Where: (p, q) is the input data, p is the benchmark rejection data that marks the corresponding benchmark endpoint data, q is the integrity requirement, and p→q means that the integrity requirement is met; the output data is the integrity judgment value;

[0087] The integrity judgment value is 1, indicating that the integrity requirements are met;

[0088] The integrity judgment value is 0, indicating that the integrity requirement is not met.

[0089] In one embodiment, the method for performing integrity analysis on the additional rejection data is the same as the method for performing integrity analysis on the baseline rejection data, that is, the baseline rejection data is replaced with the additional rejection data, the baseline segment endpoint data is replaced with the additional endpoint data, and so on, to complete the integrity analysis.

[0090] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0091] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A rapid test system for measuring error of voltage transformer low voltage test equipment based on artificial intelligence, characterized in that: Including interference analysis module, error analysis module, acquisition module and test module; The interference analysis module is used to perform environmental electromagnetic interference analysis and determine an anti-interference curve, wherein the horizontal axis of the anti-interference curve is time and the vertical axis is the anti-interference value; the anti-interference value includes 1 and 0; the anti-interference value is 1, indicating that the anti-electromagnetic interference requirement is met; the anti-interference value is 0, indicating that the anti-electromagnetic interference requirement is not met; The error analysis module is used to perform error analysis on the voltage transformer to obtain reference error analysis data and additional error analysis data, wherein the reference error analysis data includes low voltage error analysis data and load error analysis data; A baseline error test model and an additional error correction model are established based on the baseline error analysis data and the additional error analysis data respectively; The reference error test model is used to analyze the reference acquisition data to obtain the measurement test error; the additional error correction model is used to correct the measurement test error according to the additional acquisition data to obtain the reference measurement error; The acquisition module is used to acquire corresponding reference acquisition data and additional acquisition data in real time when testing the voltage transformer; and sending the baseline acquisition data and the additional acquisition data to the test module; The test module is used to test the voltage transformer, obtain an anti-interference curve, adjust the received reference acquisition data and additional acquisition data according to the anti-interference curve, and mark the adjusted reference acquisition data and additional acquisition data as first acquisition data and second acquisition data respectively; Analyzing the first collected data through a reference error test model to determine a metrological test error; The second collected data and the metrology test error are analyzed by an additional error correction model to obtain a reference metrology error, which is then displayed to a user.

2. The artificial intelligence-based voltage transformer low-voltage test equipment measurement error rapid test system according to claim 1 is characterized in that: Methods for conducting environmental electromagnetic interference analysis include: Connecting to an anti-electromagnetic interference device, and obtaining anti-interference data of the anti-electromagnetic interference device in real time when the anti-electromagnetic interference device is working; Analyzing the anti-interference data to obtain an anti-interference value of the anti-interference data; The time corresponding to the anti-interference value is identified, and an anti-interference curve is generated according to the anti-interference value at each time, wherein the horizontal axis of the anti-interference curve is the time, and the vertical axis is the anti-interference value.

3. The artificial intelligence-based voltage transformer low-voltage test equipment measurement error rapid test system according to claim 2 is characterized in that: An anti-electromagnetic unit is set in the interference analysis module, and the anti-electromagnetic unit is used to resist environmental electromagnetic interference. The anti-electromagnetic unit is used to replace the anti-electromagnetic interference device. The anti-electromagnetic unit is used to resist environmental electromagnetic interference during testing, and the anti-electromagnetic interference device is no longer connected.

4. The artificial intelligence-based voltage transformer low-voltage test equipment measurement error rapid test system according to claim 2 is characterized in that: Methods for analyzing noise-resistant data include: An anti-interference analysis model is established, and the expression of the anti-interference analysis model is: Where: (St, Y) is the input data, St represents the anti-interference data at the corresponding time, t is the subscript, indicating time; Y is the anti-interference requirement; St→Y means that the anti-interference data at the corresponding time meets the anti-interference requirement; the output data is the anti-interference value KG(St, Y), and the anti-interference value is 1 or 0; The anti-interference data and the anti-interference requirements are input into the anti-interference analysis model for analysis to obtain the anti-interference value corresponding to the anti-interference data.

5. The artificial intelligence-based voltage transformer low-voltage test equipment measurement error rapid test system according to claim 1 is characterized in that: The method for adjusting the received baseline acquisition data and the additional acquisition data according to the anti-interference curve includes: Identify the collection time corresponding to the baseline collection data and the additional collection data, and generate collection time axes of the baseline collection data and the additional collection data, respectively, marked as the baseline time axis and the additional time axis; The time period corresponding to the anti-interference value of 0 in the anti-interference curve is marked as an abnormal time period, and the abnormal time period is marked in the baseline time axis and the additional time axis. The baseline acquisition data and the additional acquisition data corresponding to the abnormal time period in the baseline time axis and the additional time axis are eliminated, and the baseline acquisition data and the additional acquisition data after the data is eliminated are adjusted for integrity and consistency to obtain the adjusted baseline acquisition data and the additional acquisition data.

6. The artificial intelligence-based voltage transformer low-voltage test equipment measurement error rapid test system according to claim 5 is characterized in that: Methods for adjusting the completeness and consistency of the baseline collected data and the additional collected data after data removal include: Mark the baseline collected data and the additional collected data after the data is eliminated as baseline eliminated data and additional eliminated data respectively; perform integrity analysis on the baseline eliminated data and the additional eliminated data respectively to obtain the baseline complete data and the additional complete data corresponding to the baseline eliminated data and the additional eliminated data respectively; Identify the acquisition time axes corresponding to the baseline complete data and the additional complete data, respectively, and mark them as the baseline complete axis and the additional complete axis; determine the consistent time period based on the baseline complete axis and the additional complete axis; The baseline complete data and the additional complete data are adjusted for consistency according to the consistent time period to obtain the adjusted baseline collection data and the additional collection data.

7. The artificial intelligence-based voltage transformer low-voltage test equipment measurement error rapid test system according to claim 6 is characterized in that: Methods for completeness analysis of baseline culling data include: Step SA1: identifying the endpoint data of the benchmark rejection data according to the abnormal time period and marking it as the benchmark endpoint data; Step SA2: Determine whether the reference endpoint data meets the integrity requirement and obtain the integrity determination result; Step SA3: when the integrity judgment result satisfies the completeness requirement, no integrity adjustment is performed on the reference endpoint data, and the mark of the reference endpoint data is cancelled; When the integrity judgment result is that the completeness requirement is not met, the reference endpoint data is discarded, new reference endpoint data is formed, and the process returns to step SA2; Step SA4: looping steps SA1 to SA3 until there is no benchmark endpoint data, marking the current benchmark-removed data as benchmark complete data.

Images