Calibration method and system of intelligent measurement switch and multi-switch synchronous measurement method

Abstract

The invention belongs to the technical field of electric power measurement, and discloses a calibration method and system of an intelligent measurement switch and a multi-switch synchronous measurement method, and the method comprises the steps: obtaining voltage and current waveform data of the intelligent measurement switch under a nonlinear load condition, and extracting a frequency spectrum feature matrix of the voltage and current waveform data; constructing an amplitude-phase coupling error model representing a nonlinear incidence relation between amplitude fluctuation and phase deviation, inputting the spectrum characteristic matrix into the amplitude-phase coupling error model, and calculating an amplitude-phase coupling error vector; performing vector decomposition on the amplitude-phase coupling error vector, and separating out an in-phase component error and an orthogonal component error; and generating a calibration compensation coefficient according to the in-phase component error and the orthogonal component error, and performing vector calibration on the intelligent measurement switch, thereby improving the calibration precision of the intelligent measurement switch.

Description

Technical Field

[0001] This invention relates to the field of power metering technology, and more specifically, to a calibration method, system, and method for synchronous measurement of multiple switches using intelligent measuring switches. Background Technology

[0002] In the manufacturing and regular maintenance of intelligent measurement switches, metrological calibration is a crucial step in ensuring their measurement accuracy. Existing calibration methods are usually based on ideal operating conditions, that is, using standard linear loads (such as pure resistors) or pure sine wave signal sources, testing the equipment error at a few fixed load points (such as 100%, 50%, and 5% of rated current), and calculating simple linear gain coefficients and phase offsets for compensation.

[0003] However, in actual operation, modern power systems have a large number of nonlinear loads (such as variable frequency air conditioners, LED lighting drivers, switching power supplies, and electric vehicle charging piles) connected to the grid, causing severe distortion of the current waveform. This results in not only abundant high-order harmonics but also frequent transient pulse impacts. Since intelligent measurement switches typically use current transformers as current sampling elements, and the core material of these transformers possesses inherent hysteresis and magnetic saturation characteristics, under nonlinear load conditions, the superposition of high-order harmonic currents and the fundamental current easily alters the working magnetic flux density of the core, causing the transformer to operate in the nonlinear region. This nonlinear response of physical characteristics causes the measurement deviation (ratio difference) and phase shift (angle difference) of the fundamental amplitude to no longer be independent variables, but rather exhibit complex amplitude-phase coupling characteristics. For example, an increase in harmonic content not only directly introduces distortion errors but also pulls the fundamental phase shift through magnetic coupling effects; the magnetic saturation effect under high current also leads to nonlinear phase angle jumps. In addition, intelligent measurement switches are usually installed outdoors or in power distribution cabinets, where the temperature difference in the working environment is large. The parameters of the internal components (especially the sampling resistor and crystal oscillator) will drift with temperature changes (i.e., temperature drift).

[0004] In summary, the limitations of existing technologies are as follows: traditional calibration methods based on linear separation models neglect the strong coupling relationship between amplitude fluctuations, phase shifts, and waveform distortion; if raw sampling data without considering environmental factors and load nonlinear characteristics is used directly for calibration, the resulting calibration parameters often fail under complex nonlinear operating conditions, leading to the phenomenon that the intelligent measuring switch is calibrated but operates out of tolerance during field operation. Therefore, this invention proposes a calibration method, system, and multi-switch synchronous measurement method for intelligent measuring switches to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a calibration method for an intelligent measuring switch, comprising: Acquire voltage and current waveform data of intelligent measuring switch under nonlinear load conditions, and extract the spectral feature matrix of voltage and current waveform data; A nonlinear correlation model representing the amplitude-phase coupling error relationship between amplitude fluctuation and phase shift is constructed. The spectral feature matrix is ​​input into the amplitude-phase coupling error model, and the amplitude-phase coupling error vector is calculated. The amplitude-phase coupling error vector is decomposed into in-phase component error and quadrature component error. Based on the in-phase component error and the quadrature component error, a calibration compensation coefficient is generated to perform vector calibration on the intelligent measurement switch.

[0006] Furthermore, the methods for acquiring voltage and current waveform data include: According to the preset test conditions, a nonlinear current containing pulse impact and harmonic interference is output through the programmable load power supply; The voltage analog signal at the input terminal and the current analog signal at the output terminal of the intelligent measuring switch are obtained through a synchronous sampling circuit, and the ambient temperature data of the intelligent measuring switch are recorded simultaneously. Oversampling and digital filtering are performed on the voltage analog signal and the current analog signal to extract the sampling point sequence under steady state; Based on the ambient temperature data, the additional temperature error value is obtained through a preset temperature drift compensation table; The sampling point sequence is pre-corrected using the temperature-added error value to obtain voltage and current waveform data containing voltage and current data sequences.

[0007] Furthermore, methods for extracting the spectral feature matrix of voltage and current waveform data include: Windowing and Fourier transform were performed on the voltage and current data sequences respectively to separate the fundamental voltage component, the fundamental current component, and their respective harmonic components. Calculate the distortion parameters of each harmonic component of voltage and current relative to their respective fundamental components, and extract the fundamental voltage amplitude, fundamental current amplitude, and fundamental voltage-current phase difference. The distortion parameters, the voltage fundamental amplitude, the current fundamental amplitude, and the voltage and current fundamental phase difference are vector-concatenated and combined. The combined data is normalized to generate a spectral feature matrix that describes the waveform distortion state.

[0008] Furthermore, methods for separating the fundamental voltage component, the fundamental current component, and their respective harmonic components include: The fundamental period of the voltage data sequence is determined by a zero-crossing detection algorithm, and the truncation window width is set based on the fundamental period. Based on the truncation window width, the voltage data sequence and the current data sequence are truncated respectively; The truncated voltage and current data sequences are weighted using window functions to obtain a weighted time-domain sequence containing both voltage-weighted and current-weighted time-domain sequences. Perform a Fast Fourier Transform on the weighted time-domain sequence to obtain the voltage spectrum sequence and the current spectrum sequence, and determine their respective spectral peak points; Polynomial interpolation algorithms are used to correct the frequency, amplitude, and phase of the peak points of the voltage spectrum sequence and the current spectrum sequence, respectively, to separate the fundamental voltage component, the fundamental current component, and their respective harmonic components.

[0009] Furthermore, the methods for constructing the amplitude-phase coupling error model include: Acquire multiple sets of test sample data of the intelligent measurement switch, and extract spectral features from the test sample data to generate a sample spectral feature matrix; Calculate the corresponding amplitude error sequence and phase error sequence based on the sample spectrum feature matrix; Using the sample spectrum feature matrix as the independent variable and the amplitude error sequence and the phase error sequence as the dependent variables, a fitted sample set is constructed, and the fitted sample set is mapped to an error distribution scatter plot. By using a surface fitting algorithm to fit and analyze the scatter plot of the error distribution, the coupling correlation function describing the amplitude-phase crossover relationship is determined, and an amplitude-phase coupling error model is constructed.

[0010] Furthermore, a surface fitting algorithm is used to perform fitting analysis on the scatter plot of the error distribution to determine the coupling correlation function describing the amplitude-phase crossover effect, including: Based on the correlation analysis between the sample spectral feature matrix and the amplitude error sequence and phase error sequence, key feature variables are selected; A higher-order polynomial is constructed as the fitting basis function, which includes the cross-product terms of the key feature variables; Calculate the Euclidean distance from each data point in the error distribution scatter plot to the fitted surface determined by the fitted basis function; The coefficient matrix of the higher-order polynomial is solved iteratively by minimizing the sum of squares of the Euclidean distances. Substituting the coefficient matrix obtained from the solution into the fitting basis function yields the coupling correlation function.

[0011] Furthermore, the construction of the amplitude-phase coupling error model characterizing the nonlinear correlation between amplitude fluctuation and phase shift also includes: The load nonlinearity index is calculated based on the sample spectrum feature matrix, and the load condition is divided into multiple load nonlinearity intervals according to the preset nonlinearity threshold. For different load nonlinearity ranges, the coupling correlation function is segmented and corresponding sub-error models are established for each segment. By integrating the sub-error models, a magnitude-phase coupling error model that is suitable for all operating conditions is obtained.

[0012] Further, based on the in-phase component error and the quadrature component error, calibration compensation coefficients are generated, including: The combined gain ratio is calculated using the in-phase component error and the quadrature component error, and an amplitude gain correction coefficient is generated. Calculate the arctangent of the ratio of the quadrature component error to the in-phase component error to generate the phase offset correction coefficient.

[0013] A multi-switch synchronous measurement method, comprising at least two intelligent measurement switches calibrated based on the aforementioned intelligent measurement switch calibration method, including: Send a synchronous measurement command to the at least two smart measurement switches. The at least two intelligent measurement switches synchronously sample their respective associated circuit branches according to a unified time reference to obtain synchronous sampling data. The synchronous sampling data is received and processed to obtain the synchronous electrical parameters of each branch.

[0014] A calibration system for an intelligent measuring switch, used to implement the calibration method for the intelligent measuring switch, includes: The spectrum analysis module is used to acquire voltage and current waveform data of the intelligent measuring switch under nonlinear load conditions and extract the spectrum feature matrix of the voltage and current waveform data. The error calculation module is used to construct an amplitude-phase coupling error model that characterizes the nonlinear correlation between amplitude fluctuation and phase shift. The spectral feature matrix is ​​input into the amplitude-phase coupling error model to calculate the amplitude-phase coupling error vector. The error vector decomposition module is used to perform vector decomposition on the amplitude-phase coupling error vector to separate the in-phase component error and the quadrature component error. The vector calibration compensation module is used to generate calibration compensation coefficients based on the in-phase component error and the quadrature component error, and to perform vector calibration on the intelligent measurement switch.

[0015] The technical effects and advantages of this invention are as follows: 1. This method solves the problem of amplitude-phase coupling error compensation under nonlinear loads. By constructing an amplitude-phase coupling error model that includes cross-product terms, this method can quantify the nonlinear correlation between amplitude fluctuations and phase shifts. Compared with traditional linear calibration methods that ignore the coupling effect between the two, this method can effectively compensate for the combined error caused by harmonic interference and magnetic saturation effect, thereby improving the measurement accuracy of intelligent measuring switches under complex nonlinear load conditions such as frequency conversion and rectification.

[0016] 2. Vector decoupling and precise correction of errors are achieved. This method decomposes the calculated amplitude-phase coupling error vector into in-phase and quadrature component errors, and generates amplitude gain correction coefficients and phase offset correction coefficients accordingly. This calibration mechanism based on vector decomposition can accurately map complex nonlinear errors to the gain and phase control parameters of digital signal processing, enabling the device to simultaneously correct ratio and angle errors, thus improving the targeting and effectiveness of calibration.

[0017] 3. Improved feature extraction accuracy and data quality. In the data processing stage, this method adopts a spectrum analysis strategy of truncation, windowing, and polynomial interpolation to effectively suppress spectral leakage and picket fence effect, and improve the extraction accuracy of fundamental amplitude, phase, and distortion parameters. At the same time, combined with the temperature drift pre-correction step, the interference of temperature factors is eliminated, ensuring that the spectral feature matrix of the input model can truly and purely reflect the physical characteristics of the load.

[0018] 4. Adaptive model matching across the entire operating range is achieved. By introducing a load nonlinearity index and adopting a piecewise modeling strategy, this method can automatically match sub-error models of different complexities based on real-time load characteristics (such as linearity). This piecewise integration approach avoids numerical oscillations that may occur in a single high-order model in the low-load region, ensuring stability under normal operating conditions. On the other hand, it ensures that the model has sufficient fitting ability under high-distortion conditions, achieving error coverage across the entire range. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the calibration method for the intelligent measuring switch of the present invention; Figure 2 This is a schematic diagram of the multi-switch synchronous measurement method of the intelligent measuring switch of the present invention; Figure 3 This is a schematic diagram of the calibration system for the intelligent measuring switch of the present invention. Detailed Implementation

[0020] 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. Example

[0021] Please see Figure 1 As shown, the calibration method for the intelligent measurement switch described in this embodiment includes: Acquire voltage and current waveform data of intelligent measuring switch under nonlinear load conditions, and extract the spectral feature matrix of voltage and current waveform data; A nonlinear correlation model representing the amplitude-phase coupling error relationship between amplitude fluctuation and phase shift is constructed. The spectral feature matrix is ​​input into the amplitude-phase coupling error model, and the amplitude-phase coupling error vector is calculated. The amplitude-phase coupling error vector is decomposed into in-phase component error and quadrature component error. Based on the in-phase component error and the quadrature component error, a calibration compensation coefficient is generated to perform vector calibration on the intelligent measurement switch.

[0022] In existing intelligent measurement switch calibration processes, standard linear loads or pure sine wave signals are typically used for testing to obtain calibration parameters. However, the obtained calibration parameters are at risk of failure under actual nonlinear operating conditions and cannot accurately reflect the true situation of amplitude-phase coupling error. To avoid the above problems, this embodiment uses the following steps to obtain voltage and current waveform data that conforms to the usage scenario: Using a high-precision programmable load power supply as a signal source, and based on preset test conditions, the power supply is programmed to output a nonlinear current containing pulse impulses and harmonic interference to simulate actual power consumption scenarios. Specifically, a nonlinear load condition is constructed by superimposing odd-order harmonics (such as the 3rd, 5th, and 7th harmonics) into the fundamental current and then superimposing periodic transient pulse signals.

[0023] A high-precision synchronous sampling circuit is used to simultaneously acquire the analog voltage signal at the input terminal and the analog current signal at the output terminal of the intelligent measuring switch; and a temperature sensor closely attached to the key metering chip of the intelligent measuring switch is used to simultaneously record the ambient temperature data of the intelligent measuring switch. The purpose of synchronous sampling is to ensure strict alignment of voltage and current data in the time domain, avoiding additional phase errors introduced by sampling time differences.

[0024] After obtaining the voltage and current analog signals, they need to be preprocessed separately. The preprocessing includes oversampling and digital filtering. In the oversampling process, a sampling rate much higher than the Nyquist frequency is used to acquire data, so as to retain more high-frequency detail information and reduce quantization noise. The digital filtering process uses a finite impulse response filter to filter out high-frequency noise points that are outside the frequency band of interest. After processing, one or more segments of stable data regions are extracted to obtain the sampling point sequence in a stable state.

[0025] Based on the recorded ambient temperature data, a pre-defined temperature drift compensation table is consulted. This table, pre-constructed based on extensive experimental data, records the gain drift and zero drift of the sampling channels at different temperatures. The corresponding temperature-related error value is obtained by looking up the table, and this value is used to perform reverse compensation (pre-correction) on the obtained sampling point sequence, thereby eliminating the influence of temperature factors on the measurement results and obtaining the final voltage and current waveform data, including voltage and current data sequences, for subsequent analysis.

[0026] Through the above steps, the actual nonlinear load condition was restored, which not only eliminated random noise and aliasing interference, but also decoupled the temperature error from the amplitude-phase coupling error. The data after temperature drift pre-correction reflects the waveform distortion characteristics caused by the nonlinear load, providing accurate data basis for the subsequent construction of a high-precision amplitude-phase coupling error model.

[0027] The acquired voltage and current waveform data are time-domain signals, with high data dimensionality and containing a large amount of redundant information. Simple time-domain waveforms are difficult to directly quantify the frequency domain distortion characteristics caused by nonlinear loads, and also difficult to directly reveal the complex coupling relationship between the fundamental amplitude, phase, and harmonic components. Directly inputting time-domain data into the error model would result in excessive computational load and difficulty in convergence. Therefore, this embodiment uses the following steps to extract the spectral feature matrix: The voltage and current waveform data are windowed, truncated, and subjected to Fourier transform to separate the fundamental component from each harmonic component; specifically, this includes the following steps: A zero-crossing detection algorithm is used to scan the voltage data sequence (because voltage signals are more stable than current signals, and zero-crossing points are more accurate). The algorithm detects the moment when the waveform data symbols flip, and the fundamental period of the voltage data sequence is determined by calculating the time difference between adjacent zero-crossing points. To minimize leakage and ensure full-cycle sampling, a truncation window width is set based on the fundamental period. This truncation window width is typically chosen to contain an integer number of fundamental periods (e.g., 5 or 10). Based on this truncation window width, both the voltage and current data sequences are truncated, discarding incomplete periodic data.

[0028] To further suppress spectral leakage, especially sidelobe interference, the truncated voltage and current data sequences are weighted separately. The window function selected is either a Blackman window or a Hanning window, which has low sidelobe characteristics. The time-domain coefficients of the window function are multiplied point-by-point with the truncated data, resulting in voltage-weighted and current-weighted time-domain sequences. This processing smooths abrupt changes at both ends of the data, allowing the signal to transition smoothly to zero at the truncation boundary.

[0029] Fast Fourier transform is performed on the voltage-weighted time-domain sequence and the current-weighted time-domain sequence respectively to convert the signal from the time domain to the frequency domain, thereby obtaining the voltage spectrum sequence and the current spectrum sequence. The current peak coefficient is then calculated based on the current spectrum sequence.

[0030] In the voltage and current spectrum sequences, the point with the largest magnitude is searched as the main peak position of the fundamental wave. Then, the peak values ​​of each harmonic are searched based on integer multiples of the fundamental frequency, thus determining the spectral peak points for each (voltage and current). The peak points obtained at this point correspond to discrete frequency indices, and due to the picket fence effect, these points may not be the actual signal peaks.

[0031] To avoid precision loss, polynomial interpolation algorithms (such as bispectral interpolation) are used to correct the frequency, amplitude, and phase of the peak points of the voltage and current spectrum sequences, respectively. Specifically, the determined peak points and their adjacent spectral points (three or more in total) are selected to construct a high-order polynomial (usually a second-order parabolic model) that approximates the true spectrum curve. By solving for the extreme points of this polynomial, the true center frequency deviation is calculated. Based on this deviation, the amplitude and phase of each component are corrected using interpolation correction formulas, thereby eliminating errors caused by the picket fence effect. Finally, based on the corrected information, the fundamental voltage component, the fundamental current component, and their corresponding harmonic components are separated.

[0032] Extract the absolute physical quantities corresponding to the fundamental wave, namely the voltage fundamental wave amplitude and the current fundamental wave amplitude, which reflect the core characteristics of the fundamental wave signal; calculate the voltage and current fundamental wave phase difference by subtracting the voltage fundamental wave absolute phase and the current fundamental wave absolute phase obtained in the interpolation correction; this phase difference reflects the current power factor angle information and is a key variable for analyzing amplitude-phase coupling.

[0033] Distortion parameters are calculated for the separated fundamental and harmonic components, namely, the harmonic content of voltage and the harmonic content of current are calculated separately to form voltage harmonic content sequences and current harmonic content sequences; at the same time, the total harmonic distortion rate of voltage and the total harmonic distortion rate of current are calculated separately; the current crest coefficient, the harmonic content of voltage, the harmonic content of current, the total harmonic distortion rate of voltage, and the total harmonic distortion rate of current together constitute the distortion parameters.

[0034] The distortion parameters, voltage fundamental amplitude, current fundamental amplitude, and voltage and current fundamental phase difference obtained from the above calculations are concatenated and combined to form an original feature set that can be used as input for subsequent models.

[0035] Because the fundamental amplitude (typically several hundred volts or tens of amperes) and the distortion parameter (typically a decimal between 0 and 1) differ significantly in magnitude, the combined data undergoes normalization to avoid this numerical difference affecting the subsequent weight allocation in the model. Normalization employs max-min normalization or Z-score standardization to map all feature data to the [0,1] or standard normal distribution range. This ultimately generates a spectral feature matrix describing the waveform distortion state. Each row of this matrix represents a waveform feature vector at a sampling time, and each column represents a spectral feature dimension.

[0036] Through the above steps, high-dimensional time-domain waveform data can be transformed into a low-dimensional spectral feature matrix with clear physical meaning. This matrix not only retains basic metrological information such as fundamental amplitude and fundamental phase, but also quantifies the influence of nonlinear load on the waveform through accurate distortion parameters and precisely calculated phase differences. In addition, since the normalization process eliminates dimensional differences, the generated feature matrix can be efficiently and stably input into the subsequent amplitude-phase coupling error model, improving the computational efficiency and generalization ability of the calibration algorithm.

[0037] Traditional calibration of electricity meters or measuring switches typically employs a component-based calibration method, assuming that amplitude error is only related to current magnitude and phase error is only related to power factor. However, under nonlinear load conditions (such as when connected to a large number of switching power supplies or frequency converters), harmonic currents can cause amplitude and phase errors to interpenetrate through the magnetic saturation characteristics of the transformer and the nonlinear response of the sampling circuit. For example, the injection of higher harmonics not only alters the amplitude reading but also causes an additional shift in the fundamental phase. If this amplitude-phase crossover effect is ignored, the calibrated equipment will still fail to meet the metering accuracy requirements under complex operating conditions. To achieve accurate calibration, a mathematical model that accurately characterizes this correlation is needed; therefore, this embodiment provides the following solution: Connect the programmable load power supply, a high-precision standard power analyzer (reference standard terminal, the accuracy level must be at least two levels higher than the device under test), and the intelligent measurement switch to the same test circuit; in the test circuit, the voltage circuit is connected in parallel to ensure that the standard meter and the intelligent measurement switch are subjected to the same voltage excitation; the current circuit is connected in series to ensure that the current signal flowing through the standard meter and the intelligent measurement switch is the same; and start the data acquisition of the standard meter and the intelligent measurement switch simultaneously through a hardware trigger signal (such as using a second pulse PPS or a synchronous trigger line).

[0038] Multiple sets of test sample data of the intelligent measuring switch under various nonlinear load conditions are acquired. For the Kth set of test data (each set of test data corresponds to a different operating condition), the spectral features of the test sample data are extracted using the methods described in the previous embodiments (i.e., truncation, FFT, interpolation correction, etc.), generating a corresponding sample spectral feature matrix. In order to construct the model input, the sample spectral feature matrix also needs to be averaged in the row dimension to obtain a sample spectral feature vector that can represent the stability characteristics of the set of operating conditions, denoted as . Simultaneously, the true value of the k-th operating condition is measured using a high-precision standard power analyzer connected in series in the circuit, and the corresponding amplitude error is calculated. With phase error (Y is the output dependent variable of the model); through the above steps, the fitted sample set can be constructed. Where M represents the number of sample data groups.

[0039] The sample spectral feature matrix (containing dimensions such as fundamental amplitude, distortion parameters, and phase difference) is aligned one-to-one with the amplitude error sequence and phase error sequence to construct a fitted sample set. To intuitively analyze the error distribution pattern, the fitted sample set is mapped to a multi-dimensional space to form an error distribution scatter plot. In this scatter plot, the spatial coordinate axes represent different spectral features (e.g., the X-axis represents the fundamental current, the Y-axis represents the total harmonic distortion rate), and the Z-axis represents the error value.

[0040] The scatter plot of the error distribution is fitted and analyzed using a surface fitting algorithm. By observing the distribution pattern of the scatter plot in the multidimensional feature space (e.g., by observing under what feature combinations the error surface exhibits drastic curling or divergence), the main driving factors causing nonlinear changes in the error are identified. Based on the results of this fitting analysis, a coupling correlation function describing the amplitude-phase cross-influence relationship is determined, thereby establishing the amplitude-phase coupling error model.

[0041] In this embodiment, the fitting analysis results show that the nonlinear error of the intelligent measurement switch is mainly caused by the magnetic saturation and hysteresis effects of the current transformer, and is modulated by phase and distortion. When the three variables of fundamental current amplitude, fundamental voltage-current phase difference, and total harmonic distortion rate of the current change simultaneously, the error values ​​of the scatter plot show significant mutual modulation characteristics (i.e., cross-influence); therefore, the fundamental amplitude, fundamental phase difference, and total harmonic distortion rate are taken as key characteristic variables.

[0042] Based on the key feature variables, a high-order polynomial containing the cross-product term of these key feature variables is constructed as the fitting basis function. To balance fitting accuracy and computational efficiency, the fitting basis function adopts a structure of "full-dimensional linear term + core cross term". The expression is as follows: ; In the formula, This represents the input sample spectral feature vector, containing n feature dimensions; This is a constant bias term used to characterize the zero-point drift error of the system; This represents a full-dimensional linear term, which contains the linear weighted sum of all n features (including key and minor variables), with s as the feature subscript. This term ensures that the model can distinguish the different linear weighted effects of different harmonics (such as the 3rd and 5th) on the error, and preserves the ability to analyze the fine structure of the spectrum. This represents the amplitude nonlinearity coefficient, used to characterize key variables. The fundamental amplitude of the current itself exhibits second-order nonlinearity. The amplitude-phase coupling coefficient is used to represent... (Current fundamental amplitude) and Second-order crossover effect between (the fundamental phase difference of voltage and current); The amplitude-distortion coupling coefficient characterizes... The modulation effect of (total harmonic distortion of current) on the measurement error of the fundamental current amplitude; It represents the phase-distortion coupling coefficient, which characterizes the interference of total harmonic distortion of the current on the measurement of the phase difference between the fundamental voltage and current. It is a three-dimensional fully coupled coefficient, which characterizes the deep coupling error under the simultaneous action of amplitude, phase and distortion, and is used to correct extreme nonlinear conditions.

[0043] Calculate the Euclidean distance from each data point in the error distribution scatter plot to the fitted surface determined by the fitted basis function. Specifically, for the fitted sample set... The Kth sample (in (Representing the two calculated error values), which are then substituted into the basis function to calculate the predicted value. ,in The Euclidean distance The formula for calculating (i.e., residual) is: .

[0044] The coefficients of the fitted basis functions are calculated by minimizing the sum of squares of the Euclidean distances; the objective function for minimizing the sum of squares of the Euclidean distances is: ;in The coefficient matrix contains all the coefficients to be determined. ,in, The transpose sign is used; the coefficient matrix of the higher-order polynomial is solved iteratively using a numerical optimization algorithm (such as the Levenberg-Marquardt algorithm or the Gauss-Newton method); during the iteration process, the algorithm is continuously adjusted. Values ​​that make the objective function The equation gradually converges to a minimum value, and the converged coefficient matrix is ​​substituted into the higher-order polynomial to obtain the coupling correlation function. This coupling correlation function is a mathematical model with defined coefficients. The microprocessor (MCU) of the intelligent measurement switch only needs to store these defined coefficients to calculate the accurate amplitude-phase coupling error value based on the real-time acquired data during runtime.

[0045] It should be noted that, This is a general placeholder. In actual model building, two independent models need to be constructed: one for predicting amplitude error and the other for predicting phase error. When constructing the amplitude error sub-model, the formula... The amplitude error of the k-th sample The coefficient matrix obtained by solving is denoted as When constructing the phase error sub-model, k in the formula refers to the phase error of the k-th sample. The coefficient matrix obtained by solving is denoted as The final amplitude-phase coupling error model actually contains two independent coupling correlation functions, which are used to output the predicted amplitude error and the predicted phase error, respectively.

[0046] To avoid the problem that a single global model cannot simultaneously ensure stability in low-load areas and accuracy in high-distortion areas, this embodiment adopts a piecewise modeling strategy. By introducing a load nonlinearity index, the complexity of the model is adaptively adjusted for different operating conditions. Specifically, this includes the following steps: The load nonlinearity index is calculated based on the sample spectral feature matrix. This index quantifies the complexity of the current operating condition and comprehensively considers the harmonic content in the frequency domain and the pulse impact characteristics in the time domain. The calculation formula is: In the formula, and These are preset weighting coefficients, with values ​​of 0.65 and 0.35, used to balance the impact of frequency domain distortion and time domain shock. The total harmonic distortion of the current; This represents the current crest coefficient.

[0047] By analyzing the curvature variation characteristics of the error distribution scatter plot, two nonlinearity thresholds were set. (0.15) and (0.6), the entire operating condition is divided into three load nonlinearity intervals: Linear interval: ; Suitable for purely resistive or slightly nonlinear loads; Weakly nonlinear intervals: For low-power loads such as household appliances; Strongly nonlinear intervals: ; Suitable for industrial rectifier or high-pulse loads; For different load nonlinearity ranges, corresponding sub-error models are established, all of which are based on the aforementioned fitting basis functions. It is obtained by retaining terms of different orders.

[0048] Linear sub-model : Within this range, all higher-order nonlinear terms are ignored, and only the linear dominant terms are retained to ensure fast computation speed and no overfitting oscillations; The expression is: .

[0049] Weak nonlinear sub-model : In this interval, the error begins to exhibit preliminary amplitude-phase coupling characteristics, requiring the introduction of a basic second-order coupling term; The expression is: .

[0050] Strongly nonlinear sub-model : In this range, the current waveform is severely distorted, requiring the use of a complete high-order model that includes fully coupled terms for calculation; The expression is: .

[0051] To ensure the continuity of the model's output when switching between different ranges, multiple sub-error models need to be integrated to obtain an amplitude-phase coupling error model that adapts to all operating conditions. Specifically, at the boundary between two sub-models, a smoothing weighting strategy is used to calculate the final output error. .by Taking the nearby transition as an example (assuming the buffer width is...) , The value is (10% to 20%), when hour, ;when hour, ; where weight The calculation formula is: .

[0052] It should be noted that the above segmented modeling process is performed independently for amplitude error and phase error respectively; that is, the final amplitude-phase coupling error model adapted to all operating conditions actually includes a set of amplitude sub-models and a set of phase sub-models.

[0053] By inputting the real-time extracted spectral feature matrix into the constructed amplitude-phase coupling error model, the current amplitude error prediction value is calculated. and phase error prediction value By combining the two vectors, the amplitude-phase coupling error vector can be obtained. This vector represents the complex deviation of the measured vector relative to the ideal true vector, encompassing both the scaling deviation of the amplitude and the rotation deviation of the phase. In the formula, It is the symbol for imaginary numbers.

[0054] To facilitate digital calibration, the synthesized amplitude-phase coupling error vector is decomposed into rectangular coordinates on the complex plane to separate the in-phase component error. Error of orthogonal components Among them, the in-phase component error refers to the projection component of the amplitude-phase coupling error vector in the direction of the standard reference vector (i.e., the true phasor measured by the standard power analyzer); the orthogonal component error refers to the projection component of the amplitude-phase coupling error vector in the direction perpendicular to the standard reference vector.

[0055] The combined gain ratio is calculated using the in-phase component error and the quadrature component error, and an amplitude gain correction coefficient is generated. ;in, The modulus represents the amplitude ratio of the measured signal to the ideal signal. Taking its reciprocal yields the amplitude gain correction coefficient used to recover the true amplitude, thereby offsetting the ratio difference of the mutual inductor.

[0056] Calculate the arctangent of the ratio of the quadrature component error to the in-phase component error to generate the phase offset correction coefficient. To compensate for the angle difference of the mutual inductors; where, . Example

[0057] Please see Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. A method for synchronous measurement of multiple switches with an intelligent measuring switch is provided, and the specific steps are as follows: Based on the intelligent measurement switch calibration method described above, after calculating the amplitude-phase coupling error vector and writing the calibration compensation coefficients, a multi-switch synchronous measurement method is used to judge the calibration results in order to verify the calibration effect and implement product quality control, thereby improving the judgment efficiency. The specific steps are as follows: The current loops of multiple smart measurement switches are connected in series to a programmable load power supply, and the voltage loops are connected in parallel to a standard voltage source to ensure consistency between the test environment and the calibration environment. The control center sends a synchronous measurement command to at least two of the smart measurement switches that have completed calibration. This command includes a unified time alignment mark to trigger each switch to start the measurement task at the same time.

[0058] Upon receiving a synchronization measurement command, the at least two intelligent measurement switches synchronously sample their respective associated circuit branches based on a unified time reference. During sampling, the metering unit inside each intelligent measurement switch calls pre-written calibration compensation coefficients—specifically, amplitude gain correction coefficients and phase offset correction coefficients generated based on in-phase component errors and quadrature component errors—to perform real-time vector correction calculations on the acquired raw voltage and current signals, thereby obtaining calibrated synchronous sampling data. Subsequently, each intelligent measurement switch uploads the processed data to the control center, which receives and parses this data to obtain the synchronous electrical parameters of each branch. These synchronous electrical parameters include key indicators such as effective voltage value, effective current value, active power, and phase angle.

[0059] While acquiring the aforementioned synchronous electrical parameters, the control center also reads the standard electrical parameter values ​​output by the high-precision standard meter at the same time, and compares and analyzes the synchronous electrical parameters reported by the intelligent measuring switch with the standard electrical parameter values. It calculates the error values ​​of the voltage, current, and power parameters of the switch under test relative to the standard values, and determines the pass / fail status of the switch according to the preset accuracy level standard. The specific determination process is as follows: The calculated error value is compared with a preset allowable error threshold. If the errors of all key electrical parameters of the intelligent measuring switch are less than or equal to the preset allowable error threshold, the intelligent measuring switch is deemed calibrated successfully, and a test pass record is generated. If the error of any electrical parameter exceeds the allowable error threshold, the intelligent measuring switch is deemed unqualified and marked as unqualified or requiring recalibration. Through the above process, a closed-loop test from nonlinear error compensation to synchronous measurement verification is achieved, ensuring that the metrological performance of the equipment leaving the factory meets the requirements. Example

[0060] Please see Figure 3 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A calibration system for an intelligent measurement switch is provided, including: The spectrum analysis module is used to acquire voltage and current waveform data of the intelligent measuring switch under nonlinear load conditions and extract the spectrum feature matrix of the voltage and current waveform data. The error calculation module is used to construct an amplitude-phase coupling error model that characterizes the nonlinear correlation between amplitude fluctuation and phase shift. The spectral feature matrix is ​​input into the amplitude-phase coupling error model to calculate the amplitude-phase coupling error vector. The error vector decomposition module is used to perform vector decomposition on the amplitude-phase coupling error vector to separate the in-phase component error and the quadrature component error. The vector calibration compensation module is used to generate calibration compensation coefficients based on the in-phase component error and the quadrature component error, and to perform vector calibration on the intelligent measurement switch.

[0061] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0062] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0063] In the description of this invention, it should be understood that the terms "first," "second," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0064] In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0065] In the description of this invention, "several" means one or more, and "a large number" means two or more.

[0066] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0067] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0068] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A calibration method for an intelligent measuring switch, characterized in that, include: Acquire voltage and current waveform data of intelligent measuring switch under nonlinear load conditions, and extract the spectral feature matrix of voltage and current waveform data; A nonlinear correlation model representing the amplitude-phase coupling error relationship between amplitude fluctuation and phase shift is constructed. The spectral feature matrix is ​​input into the amplitude-phase coupling error model, and the amplitude-phase coupling error vector is calculated. The amplitude-phase coupling error vector is decomposed into in-phase component error and quadrature component error. Based on the in-phase component error and the quadrature component error, a calibration compensation coefficient is generated to perform vector calibration on the intelligent measurement switch.

2. The method according to claim 1, characterized in that, Methods for acquiring voltage and current waveform data include: According to the preset test conditions, a nonlinear current containing pulse impact and harmonic interference is output through the programmable load power supply; The voltage analog signal at the input terminal and the current analog signal at the output terminal of the intelligent measuring switch are obtained through a synchronous sampling circuit, and the ambient temperature data of the intelligent measuring switch are recorded simultaneously. Oversampling and digital filtering are performed on the voltage analog signal and the current analog signal to extract the sampling point sequence under steady state; Based on the ambient temperature data, the additional temperature error value is obtained through a preset temperature drift compensation table; The sampling point sequence is pre-corrected using the temperature-added error value to obtain voltage and current waveform data containing voltage and current data sequences.

3. The method according to claim 2, characterized in that, Methods for extracting the spectral feature matrix of voltage and current waveform data include: Windowing and Fourier transform were performed on the voltage and current data sequences respectively to separate the fundamental voltage component, the fundamental current component, and their respective harmonic components. Calculate the distortion parameters of each harmonic component of voltage and current relative to their respective fundamental components, and extract the fundamental voltage amplitude, fundamental current amplitude, and fundamental voltage-current phase difference. The distortion parameters, the voltage fundamental amplitude, the current fundamental amplitude, and the voltage and current fundamental phase difference are vector-concatenated and combined. The combined data is normalized to generate a spectral feature matrix that describes the waveform distortion state.

4. The method according to claim 3, characterized in that, Methods for separating the fundamental voltage component, the fundamental current component, and their respective harmonic components include: The fundamental period of the voltage data sequence is determined by a zero-crossing detection algorithm, and the truncation window width is set based on the fundamental period. Based on the truncation window width, the voltage data sequence and the current data sequence are truncated respectively; The truncated voltage and current data sequences are weighted using window functions to obtain a weighted time-domain sequence containing both voltage-weighted and current-weighted time-domain sequences. Perform a Fast Fourier Transform on the weighted time-domain sequence to obtain the voltage spectrum sequence and the current spectrum sequence, and determine their respective spectral peak points; Polynomial interpolation algorithms are used to correct the frequency, amplitude, and phase of the peak points of the voltage spectrum sequence and the current spectrum sequence, respectively, to separate the fundamental voltage component, the fundamental current component, and their respective harmonic components.

5. The method according to claim 4, characterized in that, The methods for constructing the amplitude-phase coupling error model include: Acquire multiple sets of test sample data of the intelligent measurement switch, and extract spectral features from the test sample data to generate a sample spectral feature matrix; Calculate the corresponding amplitude error sequence and phase error sequence based on the sample spectrum feature matrix; Using the sample spectrum feature matrix as the independent variable and the amplitude error sequence and the phase error sequence as the dependent variables, a fitted sample set is constructed, and the fitted sample set is mapped to an error distribution scatter plot. By using a surface fitting algorithm to fit and analyze the scatter plot of the error distribution, the coupling correlation function describing the amplitude-phase crossover relationship is determined, and an amplitude-phase coupling error model is constructed.

6. The method according to claim 5, characterized in that, The error distribution scatter plot is fitted and analyzed using a surface fitting algorithm to determine the coupling correlation function describing the amplitude-phase crossover effect, including: Based on the correlation analysis between the sample spectral feature matrix and the amplitude error sequence and phase error sequence, key feature variables are selected; A higher-order polynomial is constructed as the fitting basis function, which includes the cross-product terms of the key feature variables; Calculate the Euclidean distance from each data point in the error distribution scatter plot to the fitted surface determined by the fitted basis function; The coefficient matrix of the higher-order polynomial is solved iteratively by minimizing the sum of squares of the Euclidean distances. Substituting the coefficient matrix obtained from the solution into the fitting basis function yields the coupling correlation function.

7. The method according to claim 6, characterized in that, The construction of the amplitude-phase coupling error model characterizing the nonlinear correlation between amplitude fluctuation and phase shift also includes: The load nonlinearity index is calculated based on the sample spectrum feature matrix, and the load condition is divided into multiple load nonlinearity intervals according to the preset nonlinearity threshold. For different load nonlinearity ranges, the coupling correlation function is segmented and corresponding sub-error models are established for each segment. By integrating the sub-error models, a magnitude-phase coupling error model that is suitable for all operating conditions is obtained.

8. The method according to claim 7, characterized in that, Based on the in-phase component error and the quadrature component error, calibration compensation coefficients are generated, including: The combined gain ratio is calculated using the in-phase component error and the quadrature component error, and an amplitude gain correction coefficient is generated. Calculate the arctangent of the ratio of the quadrature component error to the in-phase component error to generate the phase offset correction coefficient.

9. A method for synchronous measurement of multiple switches, characterized in that, At least two intelligent measurement switches calibrated based on the calibration method of the intelligent measurement switch according to claim 1, the method comprising: Send a synchronous measurement command to the at least two smart measurement switches. The at least two intelligent measurement switches synchronously sample their respective associated circuit branches according to a unified time reference to obtain synchronous sampling data. The synchronous sampling data is received and processed to obtain the synchronous electrical parameters of each branch.

10. A calibration system for an intelligent measuring switch, used to implement the method according to any one of claims 1 to 8, characterized in that, include: The spectrum analysis module is used to acquire voltage and current waveform data of the intelligent measuring switch under nonlinear load conditions and extract the spectrum feature matrix of the voltage and current waveform data. The error calculation module is used to construct an amplitude-phase coupling error model that characterizes the nonlinear correlation between amplitude fluctuation and phase shift. The spectral feature matrix is ​​input into the amplitude-phase coupling error model to calculate the amplitude-phase coupling error vector. The error vector decomposition module is used to perform vector decomposition on the amplitude-phase coupling error vector to separate the in-phase component error and the quadrature component error. The vector calibration compensation module is used to generate calibration compensation coefficients based on the in-phase component error and the quadrature component error, and to perform vector calibration on the intelligent measurement switch.

Images