A data processing method for testing a current transformer

Abstract

The present application belongs to the technical field of mutual inductor verification, and particularly relates to a data processing method for current transformer testing, to solve the technical problem that the existing verification method cannot truly reflect the comprehensive error of the mutual inductor, comprising the following steps: S1, obtaining current waveform data of the measured current transformer; S2, deducting the exponential decay non-periodic component from the original current waveform data to obtain corrected current waveform data; S3, reconstructing to obtain denoised current waveform data; S4, obtaining the amplitude and phase angle of the fundamental wave and set harmonic of the primary side and secondary side currents; S5, calculating the initial ratio difference and initial angle difference of the current transformer; S6, generating a ratio difference compensation vector and an angle difference compensation vector; S7, performing vector summation of the initial ratio difference and initial angle difference with the ratio difference compensation vector and the angle difference compensation vector respectively to obtain the final ratio difference and angle difference of the current transformer. The test accuracy and reliability of the ratio difference and angle difference of the current transformer are improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of mutual inductor verification, and particularly relates to a data processing method for current transformer testing. BACKGROUND

[0002] The current transformer is a key device for measurement, protection and monitoring in the power system, which converts the large current on the primary side into small current on the secondary side according to a predetermined ratio for the use of secondary devices such as instruments and relay protection. The accuracy of the current transformer measurement performance is directly related to the power metering, fault diagnosis and safe and stable operation of the power grid. Therefore, it is crucial to periodically and accurately verify the current transformer on site. The traditional current transformer verification method mainly relies on standard current transformers and transformer calibrators, and measures by comparison. This method usually needs to be carried out under standard power frequency sinusoidal steady state conditions, and the test system is bulky, requiring a large power boost source, and the field wiring and operation are complex, with high work intensity and low efficiency. This method cannot effectively simulate and evaluate the dynamic performance of the current transformer under complex working conditions such as transient faults and harmonic pollution that the current transformer may encounter in the actual power grid, and the test results are difficult to fully reflect the real working characteristics of the transformer.

[0003] With the rapid development of digital signal processing technology and computer technology, a current transformer testing method based on digital sampling has emerged. This method acquires the instantaneous value waveform of the primary and secondary side currents, analyzes them using algorithms, and obtains the ratio error and phase error. However, the existing digital processing method still has many technical bottlenecks. First, when testing transient characteristics, the primary side current often contains a non-periodic component that decays exponentially. If not processed, it will directly interfere with the subsequent Fourier transform and other spectral analysis algorithms, causing spectral leakage and affecting the extraction accuracy of the fundamental and harmonic parameters. Second, the collected signals inevitably mix random noise such as white noise and power frequency interference, reducing the signal-to-noise ratio and affecting the accuracy of the measurement. In addition, the classic fast Fourier transform (FFT) algorithm has the problems of fence effect and spectral leakage, and its frequency resolution is limited by the sampling time and the number of points, making it difficult to perform fine analysis of the signal spectrum with high resolution, resulting in errors in amplitude and phase angle calculation. More importantly, most methods only consider the fundamental component to calculate the error, ignoring the impact of the nonlinear characteristics of the transformer core. In fact, under harmonic or transient high-current conditions, each harmonic component will affect the main flux of the core, producing additional error components. Simply calculating the ratio error and phase error based on the fundamental component cannot truly reflect the comprehensive error of the transformer. Therefore, how to accurately compensate for the nonlinear error caused by harmonics is a core technical problem in improving the accuracy of transformer testing. SUMMARY

[0004] The application provides a data processing method for current transformer testing, to solve the technical problem that the existing verification method cannot truly reflect the comprehensive error of the current transformer.

[0005] To solve the above problems, the data processing method for current transformer testing provided by the application adopts the following technical scheme: a data processing method for current transformer testing, comprising the following steps:

[0006] S1, acquiring current waveform data of the measured current transformer under transient operating conditions or steady operating conditions, the current waveform data being discrete sampling results of the primary side and the secondary side;

[0007] S2, fitting an exponential decay non-periodic component in the original current waveform data by using the least square method, and then deducting the exponential decay non-periodic component from the original current waveform data to obtain corrected current waveform data;

[0008] S3, performing multi-layer decomposition on the corrected current waveform data by using symlet wavelet, and obtaining denoised current waveform data through threshold processing of wavelet coefficients based on Stein unbiased risk estimation and reconstruction;

[0009] S4, performing frequency spectrum analysis on the denoised current waveform data by using Chirp Z transform to acquire the amplitude and phase angle of the fundamental wave and the set harmonic of the primary side and the secondary side currents;

[0010] S5, calculating the initial ratio difference and the initial angle difference of the current transformer according to the amplitude and phase angle of the fundamental wave of the primary side and the secondary side currents;

[0011] S6, generating a ratio difference compensation vector and an angle difference compensation vector according to a preset harmonic flux compensation model, taking the amplitude and phase angle of the fundamental wave and the amplitudes and phase angles of the 3rd, 5th and 7th harmonics of the secondary side current as inputs;

[0012] S7, performing vector summation on the initial ratio difference and the initial angle difference and the ratio difference compensation vector and the angle difference compensation vector respectively to obtain the final ratio difference and the final angle difference of the current transformer.

[0013] Further, in S1, the current signals of the primary side and the secondary side are synchronously collected by a synchronous data collection card at a set sampling frequency, and the collection is continued for several power frequency periods.

[0014] Further, S2 comprises the following steps:

[0015] establishing a mathematical model for the original current waveform data is expressed as: ;

[0016] wherein, , , The sine amplitude, angular frequency, and initial phase of the sinusoidal component; , The initial amplitude and time constant of the exponentially decaying aperiodic component; For the first The time of each sampling point;

[0017] The parameters are obtained by iterative solution using the nonlinear least squares method. and The optimal estimate is the one that minimizes the sum of squared errors between the current waveform data output by the model and the actual current waveform data.

[0018] From the raw current waveform data The exponentially decaying aperiodic component obtained after subtracting the fitted component The corrected current waveform data is obtained.

[0019] Furthermore, the wavelet decomposition process uses the sym8 wavelet as the wavelet basis function to perform a 5-level decomposition on the corrected current waveform data, obtaining the approximation coefficients and detail coefficients at each level.

[0020] Furthermore, threshold processing and reconstruction include the following steps:

[0021] For the detail coefficients of layers 1 to 5, the optimal threshold for each layer is adaptively calculated based on the Stan unbiased risk estimation criterion.

[0022] A soft thresholding function is used to denoise the detail coefficients of each layer;

[0023] The processed detail coefficients of each layer are reconstructed using wavelet reconstruction with the approximation coefficients of the 5th layer to obtain the denoised waveform data.

[0024] Furthermore, high-resolution spectral analysis is performed using the Chirp Z-transform, including the following steps:

[0025] The nominal frequencies of the fundamental frequency and each harmonic. Set respectively with A narrowband analysis frequency range with a preset width, centered on a specific frequency.

[0026] Within the narrowband analysis frequency range, Chirp Z-transform is used for spectrum refinement calculation to obtain a high-resolution discrete spectrum;

[0027] In a discrete spectrum, the amplitude and phase angle of the fundamental wave and the set second harmonic are solved by performing quadratic interpolation on the peak points and their adjacent points.

[0028] Furthermore, in S5, the initial ratio difference Calculate using the following formula: ;

[0029] wherein, is a nominal variation, , are the fundamental amplitudes of the primary and secondary currents respectively;

[0030] the initial angle difference is calculated according to the following formula: ;

[0031] wherein, , are the fundamental phase angles of the primary and secondary currents respectively.

[0032] Further, according to the preset harmonic flux compensation model, the ratio difference compensation vector and the angle difference compensation vector are calculated and generated, including the following steps: taking the fundamental amplitudes of the secondary currents and the amplitudes and phase angles of the 3rd, 5th and 7th harmonics as inputs, the ratio difference compensation amount and the angle difference compensation amount are calculated according to the following formula: ; ;

[0033] wherein, is the fundamental amplitude of the secondary current; , , are the amplitudes of the 3rd, 5th and 7th harmonics of the secondary side respectively; , , are the phase angles of the 3rd, 5th and 7th harmonics of the secondary side relative to the fundamental wave respectively; , , are the preset proportional coefficients of the 3rd, 5th and 7th harmonics in the harmonic flux compensation model for ratio difference compensation respectively; , , are the preset proportional coefficients of the 3rd, 5th and 7th harmonics in the harmonic flux compensation model for angle difference compensation respectively.

[0034] Further, the coefficients in the harmonic flux compensation model are obtained by the following steps:

[0035] A standard current containing a fundamental wave and a set of harmonic components is applied to the measured current transformer, and the actual ratio difference and angle difference generated thereby are measured synchronously by using a standard instrument;

[0036] The amplitudes and phase angles of each harmonic of the secondary side of the measured current transformer are fitted with the actual errors measured by the standard instrument by using a multiple linear regression algorithm, and the coefficients that minimize the difference between the compensation amounts calculated by the harmonic flux compensation model and the actual errors are solved.

[0037] Furthermore, the calculated initial ratio difference is taken as the real part, and the initial angular difference is converted into radians and taken as the imaginary part, thus forming the initial error complex vector;

[0038] The ratio difference compensation vector is taken as the real part, and the angle difference compensation vector is converted into radians and taken as the imaginary part. These are added to the real and imaginary parts of the initial error complex vector respectively, and the complex vector summation operation is performed to obtain the final error vector.

[0039] The real part of the error vector is the final ratio difference after compensation, and the imaginary part is the final angle difference after compensation.

[0040] The beneficial effects are as follows: Compared with existing technologies, the data processing method proposed in this invention removes non-periodic components using the least squares method and combines wavelet and Stan unbiased risk estimation thresholds for denoising, thus purifying the original sampled signal and laying a high-precision foundation for subsequent analysis. Secondly, by using Chirp Z-transform instead of traditional FFT, high-resolution extraction of fundamental and harmonic parameters is achieved, overcoming the calculation bias caused by spectral leakage and the picket fence effect. This invention establishes a harmonic flux linkage compensation model, using the second-side harmonic as a key input to compensate for errors introduced by core nonlinearity, solving the problem that existing technologies cannot accurately evaluate the performance of current transformers under harmonic or transient conditions by relying solely on fundamental wave analysis. This method improves the accuracy and reliability of current transformer ratio and angle error testing through precise signal processing and accurate error compensation. Attached Figure Description

[0041] Figure 1 This is a flowchart of a data processing method for testing current transformers. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Those skilled in the art should understand that the embodiments described below are only some, not all, of the embodiments disclosed. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0043] An embodiment of the data processing method for testing current transformers provided by this invention:

[0044] like Figure 1 As shown, the data processing method for current transformer testing includes the following steps:

[0045] S1: Obtain the current waveform data of the current transformer under test under transient or steady-state conditions. The current waveform data is the discrete sampling result of the primary and secondary sides.

[0046] In an optional embodiment, a synchronous data acquisition card is used to synchronously acquire the current signals of the primary and secondary sides at a set sampling frequency (e.g., 10kHz) for several power frequency cycles to ensure that the acquired current waveform data is complete and the phase relationship is accurate.

[0047] S2. The least squares method is used to fit the exponentially decaying aperiodic component in the original current waveform data, and then the exponentially decaying aperiodic component is subtracted from the original current waveform data to obtain the corrected current waveform data.

[0048] In an optional embodiment, S2 includes the following steps:

[0049] Establish the original current waveform data , The mathematical model is expressed as: ;

[0050] in, , , The sine amplitude, angular frequency, and initial phase of the sinusoidal component; , The initial amplitude and time constant of the exponentially decaying aperiodic component; For the first The time of each sampling point;

[0051] The parameters are obtained by iterative solution using the nonlinear least squares method. and The optimal estimate is the one that minimizes the sum of squared errors between the current waveform data output by the model and the actual current waveform data.

[0052] From the raw current waveform data The exponentially decaying aperiodic component obtained after subtracting the fitted component The corrected current waveform data is obtained.

[0053] Specifically, when a short-circuit fault occurs in a power system, the fault current often contains a DC component that decays exponentially. This DC component interferes with subsequent Fourier analysis, leading to inaccurate harmonic measurements. Therefore, a mathematical model needs to be established that can simultaneously describe both the AC sinusoidal component and the DC decaying component. For example, at a sampling point... current value It is modeled as the sum of an ideal sine wave and an exponentially decaying term.

[0054] The series of current data collected were processed using the nonlinear least squares method. , ... Fitting to the mathematical model. Nonlinear least squares method through iterative calculation, constantly adjusting parameters (initial amplitude) and parameters (time constant), the goal is to minimize the sum of squared errors between the model calculated values and the actual measured values of all sampling points. For example, after calculation, the optimal initial amplitude may be equal to 50A, and the time constant is equal to 20ms. According to these optimal parameters, the value of the exponential decay aperiodic component at each sampling time is calculated and subtracted from the original current waveform data, resulting in a pure AC waveform data, laying the foundation for subsequent accurate wavelet analysis and spectral analysis.

[0055] S3, using symlet wavelet to decompose the corrected current waveform data in multiple layers, and through threshold processing of wavelet coefficients based on Stein unbiased risk estimation, the denoised current waveform data is obtained by reconstruction.

[0056] In an optional embodiment, the wavelet decomposition part uses sym8 wavelet as the wavelet basis function to decompose the corrected current waveform data in 5 layers to obtain the approximation coefficients and detail coefficients of each level.

[0057] Specifically, the sym8 wavelet basis function is selected because it has good symmetry and compact support, which can produce smaller phase distortion and error when decomposing and reconstructing signals, and is suitable for processing transient signals in power systems. Wavelet decomposition of the current waveform data after subtraction of the aperiodic component is actually passing the current waveform data through a series of high-pass filters and low-pass filters, and the low-pass filter is used to obtain the approximation coefficient A, and the high-pass filter is used to obtain the detail coefficient D.

[0058] The 5-layer decomposition of the current waveform data is essentially separating the current waveform data into different frequency bands. Specifically, if the sampling frequency of the original current waveform data is 12.8kHz, the detail coefficient D1 obtained by the first layer decomposition corresponds to the highest frequency noise of 3.2kHz-6.4kHz. The detail coefficient D2 obtained by the second layer decomposition corresponds to the highest frequency noise of 1.6kHz-3.2kHz, and so on. The detail coefficient D5 of the fifth layer corresponds to the highest frequency noise of 200Hz-400Hz. The approximation coefficient A5 of the fifth layer corresponds to the low frequency main component of 0-200Hz, which contains the key information of the fundamental and harmonic. Through such hierarchical mode, different frequency band noises are clearly separated, which facilitates subsequent fine noise processing for each frequency band.

[0059] In an optional embodiment, the threshold processing and reconstruction include the following steps:

[0060] The optimal threshold of each layer is adaptively calculated based on Stein's unbiased risk estimate (SURE) criterion for the detail coefficients of the 1st to 5th layers;

[0061] The soft threshold function is used to denoise the detail coefficients of each layer;

[0062] The processed detail coefficients of each layer and the approximation coefficients of the 5th layer are reconstructed by wavelet to obtain the denoised waveform data.

[0063] Specifically, in order to effectively remove noise while maximizing the preservation of useful signals, an adaptive threshold determination method is used. The Stein's unbiased risk estimate (SURE) criterion can automatically calculate the optimal threshold according to the data characteristics of each layer of detail coefficients, without the need to know the noise intensity in advance. For example, for the 1st layer of detail coefficients containing a large amount of Gaussian white noise, the SURE criterion will calculate a higher optimal threshold (such as 0.9); for the 4th layer of detail coefficients with less noise, a lower optimal threshold (such as 0.15) will be obtained.

[0064] After determining the optimal threshold of each layer, the soft threshold function is used for processing. The soft threshold function will set the coefficients with absolute values less than the optimal threshold to zero, and shrink the coefficients with absolute values greater than the optimal threshold to zero in the direction of zero by a threshold amount. Compared with the hard threshold method of directly setting to zero, this method can make the reconstructed signal smoother and avoid additional oscillation. The detail coefficients (D1 to D5) of all levels after threshold processing are combined with the 5th layer approximation coefficients A5 which contain the main signal information, and the final pure current waveform data is reconstructed by wavelet inverse transform.

[0065] S4, the denoised current waveform data is analyzed by Chirp Z transform to obtain the amplitude and phase angle of the fundamental and set harmonics of the primary and secondary currents.

[0066] In an optional embodiment, high-resolution spectral analysis is performed using Chirp Z transform, including the following steps:

[0067] For the nominal frequency of the fundamental and each harmonic , a narrowband analysis frequency range with a preset width is set around ;

[0068] Within the narrowband analysis frequency range, Chirp Z transform is used for spectral refinement calculation to obtain high-resolution discrete spectrum;

[0069] In the discrete spectrum, the amplitude and phase angle of the fundamental and set harmonics are solved by quadratic interpolation of the discrete spectrum peak points and their adjacent points.

[0070] Specifically, the frequency resolution of a conventional Fourier transform (FFT) is limited by the sampling length and the number of points, and cannot accurately capture harmonics deviating from the nominal frequency. A Chirp Z transform (CZT) is used to realize local amplification of the frequency spectrum. For example, for a fundamental wave of a nominal 50 Hz, a narrowband analysis frequency range of 49 Hz to 51 Hz can be set. The CZT can calculate 1000 frequency spectrum points within this narrowband analysis frequency range of only 2 Hz, so that the frequency resolution reaches 0.002 Hz, which is much higher than the resolution provided by the FFT, thereby enabling accurate identification of the fundamental frequency deviation caused by power grid fluctuations.

[0071] Even after CZT refinement, the real frequency spectrum peak value can fall between two calculated frequency points. To further improve accuracy, a quadratic interpolation technique is used. When a peak point is found in the high-resolution frequency spectrum calculated by the CZT, three points are taken, including the peak point and its left and right adjacent points. A parabola can be fitted through the three points, and the frequency, amplitude, and phase angle corresponding to the vertex of the parabola are considered to be the most accurate estimates of the harmonic component. For example, if the CZT frequency spectrum peak value is at 50.014 Hz, a more accurate frequency of 50.0146 Hz can be obtained through quadratic interpolation, as well as the amplitude and phase angle corresponding to 50.0146 Hz.

[0072] S5, according to the fundamental wave amplitude and phase angle of the primary side and secondary side currents, the initial ratio difference and the initial angle difference of the current transformer are calculated.

[0073] In an optional embodiment, the initial ratio difference is calculated according to the following formula: ;

[0074] wherein, is the nominal change, , are the fundamental wave amplitudes of the primary side and secondary side currents, respectively.

[0075] The initial angle difference is calculated according to the following formula: ;

[0076] wherein, , are the fundamental wave phase angles of the primary side and secondary side currents, respectively.

[0077] S6, according to the preset harmonic flux compensation model, the fundamental wave amplitude and the amplitudes and phase angles of the 3rd, 5th, and 7th harmonics of the secondary side current are taken as inputs to generate a ratio difference compensation vector and an angle difference compensation vector.

[0078] In an alternative embodiment, the ratio difference compensation vector and the angle difference compensation vector are calculated according to a preset harmonic flux compensation model, comprising the following steps: taking the fundamental amplitude of the secondary side current , and the amplitudes and phase angles of the 3rd, 5th and 7th harmonics as inputs, the ratio difference compensation amount and the angle difference compensation amount are calculated by the following formulas: ; ;

[0079] wherein, is the fundamental amplitude of the secondary side current; , , are the amplitudes of the 3rd, 5th and 7th harmonics of the secondary side current, respectively; , , are the phase angles of the 3rd, 5th and 7th harmonics of the secondary side current relative to the fundamental, respectively; , , are the preset proportional coefficients of the 3rd, 5th and 7th harmonics in the harmonic flux compensation model for ratio difference compensation, respectively; , , are the preset proportional coefficients of the 3rd, 5th and 7th harmonics in the harmonic flux compensation model for angle difference compensation, respectively.

[0080] Specifically, the measurement error of the current transformer is mainly caused by the nonlinear magnetization characteristics of the core, and the presence of harmonic components will exacerbate this nonlinearity, resulting in increased error. The harmonic flux compensation model aims to quantify the additional error introduced by the 3rd, 5th and 7th harmonic components. The harmonic flux compensation model decomposes the effect of these harmonic components on the total flux into two parts: one part is in phase with the fundamental flux, corresponding to the cause of the ratio difference error; the other part is orthogonal to the fundamental flux, corresponding to the cause of the angle difference error.

[0081] For example, after frequency spectrum analysis, the secondary side fundamental current is 5A, the 3rd harmonic is 0.4A and its phase angle relative to the fundamental is 45°. At the same time, it is assumed that the preset compensation coefficients are 0.01, 0.008. Then, considering only the 3rd harmonic, the ratio difference compensation component it produces is: ; the angle difference compensation component is: .

[0082] Linearly superimposing the compensation components produced by the 3rd, 5th and 7th harmonics, respectively, the total ratio difference compensation amount and the angle difference compensation amount , for correcting the measurement results of a mutual inductor.

[0083] In an optional embodiment, the coefficients in the harmonic flux compensation model are obtained by calibration through the following steps:

[0084] A standard current containing fundamental and set harmonic components is applied to the current transformer under test, and the actual percentage error and phase error generated thereby are measured synchronously using a standard instrument;

[0085] The amplitudes and phases of the harmonics on the secondary side of the current transformer under test are fitted with the actual errors measured by the standard instrument using a multiple linear regression algorithm, to obtain the coefficients that minimize the difference between the compensation amount calculated by the harmonic flux compensation model and the actual error.

[0086] Specifically, the coefficients in the harmonic flux compensation model , are specific to each current transformer and need to be calibrated through experiments. During calibration, a programmable high-power current source is used to inject an accurately known current signal into the current transformer under test. For example, a composite current of a 50 Hz fundamental wave superimposed with 10% 3rd harmonic and 5% 5th harmonic can be injected. At the same time, a standard power quality analyzer or transformer calibrator accurately measures the percentage error and phase error actually generated by the current transformer under this working condition, for example, the measured percentage error is 0.1% and the phase error is 5 ′ .

[0087] A set of rich experimental data is obtained by changing the amplitude and phase of the injected harmonics and performing multiple measurements. Each set of data contains the amplitude and phase information of each harmonic and the corresponding measured error. These data are used as input, and a multiple linear regression algorithm is applied. The goal of the multiple linear regression algorithm is to solve a set of optimal coefficients , , , , , so that the overall deviation between the theoretical error calculated by substituting the harmonic data into the compensation formula and the error values measured at all experimental points is minimized. The set of coefficients thus obtained is then fixed in the measuring device for online real-time compensation.

[0088] S7, the initial percentage error and the initial phase error are respectively vector-summed with the percentage error compensation vector and the phase error compensation vector to obtain the final percentage error and phase error of the current transformer.

[0089] In an alternative embodiment, the calculated initial ratio difference is taken as the real part, and the initial angle difference converted to radians is taken as the imaginary part, to form an initial error complex vector. The ratio difference compensation vector is taken as the real part, and the angle difference compensation vector converted to radians is taken as the imaginary part, and added to the real and imaginary parts, respectively, of the initial error complex vector to perform a complex vector summation operation to obtain a final error vector. The real part of the error vector is the final ratio difference after compensation, and the imaginary part is the final angle difference after compensation.

[0090] Also in the description of the present specification, the meaning of "a plurality of" is at least two, for example, two, three or more, etc., unless specifically defined otherwise.

Claims

1. A data processing method for testing a current transformer, characterized by, The method comprises the following steps: S1, obtaining current waveform data of the measured current transformer under transient or steady state conditions, the current waveform data being discrete sampling results of the primary side and the secondary side; S2, fitting the exponential decay non-periodic component in the original current waveform data by using the least square method, and then deducting the exponential decay non-periodic component from the original current waveform data to obtain corrected current waveform data; S3, decomposing the corrected current waveform data by using symlet wavelet in multiple layers, and processing the wavelet coefficients by threshold value based on Stein unbiased risk estimation, and then reconstructing to obtain denoised current waveform data; S4, performing frequency spectrum analysis on the denoised current waveform data by using Chirp Z transform to obtain the amplitude and phase angle of the fundamental wave and the set harmonic of the primary side and the secondary side currents; S5, calculating the initial ratio difference and the initial angle difference of the current transformer according to the amplitude and phase angle of the fundamental wave of the primary side and the secondary side currents; S6, according to the preset harmonic magnetic chain compensation model, the fundamental amplitude of the secondary side current and the amplitude and phase angle of the 3rd, 5th and 7th harmonics are taken as inputs, and the ratio difference compensation amount is calculated through the following formula And the angle difference compensation amount : ; ； wherein, is the fundamental amplitude of the secondary side current; , , are the amplitudes of the 3rd, 5th, 7th harmonics of the secondary side, respectively; , , are the phase angles of the 3rd, 5th, 7th harmonics of the secondary side relative to the fundamental, respectively; , , are the preset proportional coefficients of the 3rd, 5th, 7th harmonics in the harmonic flux compensation model for the ratio difference compensation, respectively; , , are the preset proportional coefficients of the 3rd, 5th, 7th harmonics in the harmonic flux compensation model for the angle difference compensation, respectively, to generate the ratio difference compensation vector and the angle difference compensation vector; S7, the initial ratio difference and the initial angle difference are respectively vector summed with the ratio difference compensation vector and the angle difference compensation vector to obtain the final current transformer ratio difference and angle difference.

2. The data processing method for testing a current transformer according to claim 1, characterized by, In S1, the current signals of the primary side and the secondary side are synchronously collected by a synchronous data acquisition card at a set sampling frequency, and the collection is continued for several power frequency periods.

3. The data processing method for testing a current transformer according to claim 1, characterized by, S2 comprises the following steps: establishing a mathematical model of the raw current waveform data is expressed as: ; wherein , , is the sinus amplitude, the angular frequency, the initial phase of the sinus component; , is the initial amplitude, the time constant of the exponentially decaying aperiodic component; is the time of the th sampling point; The parameters are obtained by iterative solution using the nonlinear least squares method. and The optimal estimate is the one that minimizes the sum of squared errors between the current waveform data output by the model and the actual current waveform data. The corrected current waveform data is obtained from the original current waveform data by subtracting the fitted exponential decay aperiodic component from the original current waveform data. The corrected current waveform data is obtained from the original current waveform data by subtracting the fitted exponential decay aperiodic component from the original current waveform data. The corrected current waveform data is obtained from the original current waveform data by subtracting the fitted exponential decay aperiod 4. The data processing method for testing a current transformer according to claim 1, characterized by, In the wavelet decomposition link, sym8 wavelet is used as the wavelet base function to decompose the corrected current waveform data in 5 layers to obtain the approximation coefficients and the detail coefficients of each layer.

5. The data processing method for testing a current transformer according to claim 4, characterized in that, The threshold value processing and reconstruction comprises the following steps: For the detail coefficients of the 1st layer to the 5th layer, the optimal threshold value of each layer is adaptively calculated based on the Stein unbiased risk estimation criterion; The soft threshold function is used to denoise the detail coefficients of each layer; The processed detail coefficients of each layer and the approximation coefficients of the 5th layer are reconstructed by wavelet to obtain the denoised waveform data.

6. The data processing method for testing the current transformer according to claim 1, wherein high-resolution frequency spectrum analysis is performed by using Chirp Z transform, comprising the following steps: In the narrowband analysis frequency range, Chirp Z transform is used to perform frequency spectrum refinement calculation to obtain high-resolution discrete frequency spectrum; The nominal frequencies of the fundamental frequency and each harmonic. Set respectively with A narrowband analysis frequency range with a preset width, centered on a specific frequency. In the discrete frequency spectrum, the amplitude and phase angle of the fundamental wave and the set harmonic are solved by performing quadratic interpolation on the peak points and their adjacent points in the discrete frequency spectrum. The coefficients in the harmonic flux compensation model are obtained by the following steps:

7. The data processing method for testing a current transformer according to claim 1, characterized by, In S5, the initial ratio difference The following formula is used for calculation: ; wherein is a nominal variation, , are the fundamental amplitudes of the primary and secondary currents, respectively; initial angle difference The following equation is used to calculate: ; wherein , are the fundamental phase angles of the primary and secondary currents, respectively.

8. The data processing method for testing a current transformer according to claim 1, characterized by, A standard current containing the fundamental wave and the set harmonic components is applied to the measured current transformer, and the actual ratio difference and angle difference generated thereby are synchronously measured by using a standard instrument; The amplitude and phase angle of each harmonic of the secondary side of the measured current transformer are fitted with the actual error measured by the standard instrument by using a multiple linear regression algorithm to obtain the coefficients that minimize the difference between the compensation amount calculated by the harmonic flux compensation model and the actual error. The calculated initial ratio difference is taken as the real part, and the initial angle difference is converted into radians and taken as the imaginary part to form an initial error complex vector; 9. The data processing method for testing a current transformer according to any one of claims 1 to 8, characterized in that, The ratio difference compensation vector is taken as the real part, and the angle difference compensation vector is converted into radians and taken as the imaginary part, which are added to the real part and the imaginary part of the initial error complex vector respectively to perform complex vector summation operation to obtain the final error vector; The real part of the error vector is the final ratio difference after compensation, and the imaginary part is the final angle difference after compensation. ​

Images