An automatic compensation calibration method for electric energy metering error of an electric energy meter

Abstract

The present application belongs to the field of electric energy meter, and particularly relates to an automatic compensation calibration method for electric energy metering error of an electric energy meter, aiming at solving the technical problems that the CORDIC algorithm is difficult to balance efficiency and accuracy, and the existing compensation method is difficult to correct multi-dimensional nonlinear error under wide temperature and wide current, comprising the following steps: S1, outputting uncorrected x-axis component and final cumulative phase angle, and the improved CORDIC iteration operation comprises M times of rough rotation iteration and P times of fine rotation iteration; S2, performing amplitude correction on the uncorrected x-axis component to obtain corrected instantaneous current amplitude; S3, reconstructing the corrected instantaneous current amplitude and the final cumulative phase angle to obtain corrected instantaneous current value, multiplying the corrected instantaneous current value and instantaneous voltage value to obtain instantaneous active power, and performing cumulative integration on the instantaneous active power to obtain compensated and calibrated electric energy value. The present application can ensure that the electric energy meter maintains high-precision metering performance in a wide current range and temperature change range.

Description

Technical Field

[0001] This invention belongs to the field of electricity meters, and particularly relates to an automatic compensation and calibration method for electricity meter measurement errors. Background Technology

[0002] Electricity meters are metering devices used for settlement in power systems, and their accuracy directly affects the interests of both power suppliers and consumers. During operation, due to the non-ideal characteristics of current transformers, shunts, and analog-to-digital converters, electricity meters are prone to metering deviations under different operating conditions. Modern electronic electricity meters generally employ digital signal processing technology to analyze synchronously acquired grid voltage and line current to calculate parameters such as active power, reactive power, and energy. Commonly used methods include Fast Fourier Transform (FFT) and the Coordinate Rotation Digital Computation (CORDIC) algorithm. FFT can accurately extract signal amplitude and phase, but it requires a large number of complex multiplications, resulting in a large computational load and high processor performance requirements. In contrast, the CORDIC algorithm can complete the conversion between rectangular and polar coordinates through shifting and addition / subtraction operations, thus efficiently solving for the instantaneous amplitude and phase of the signal, and has the advantages of high computational efficiency and ease of hardware implementation.

[0003] However, the CORDIC algorithm suffers from a trade-off between iterative convergence speed and computational accuracy. To achieve high accuracy, a large number of iterations are typically required, increasing computational latency and impacting real-time metering. To address this, several compensation methods are employed: some use fixed linear gain and offset for correction, but these struggle to handle the nonlinear characteristics of sensors at different current levels and operating temperatures; others use lookup tables for compensation, but often only correct for the influence of a single variable, or the compensation model is overly simplified, failing to accurately correct the nonlinear errors caused by the combined effects of multiple variables such as current and temperature. Because sensor errors exhibit strong nonlinearity and multivariate coupling characteristics over a wide temperature and dynamic current range, and existing compensation methods are either linear models or only address a single variable, they struggle to accurately fit the true error surface. Therefore, the compensation effect deteriorates significantly when deviating from the rated operating point, leading to decreased metering accuracy. Improving efficiency and accuracy is a pressing technical problem in the current electricity meter industry. Summary of the Invention

[0004] This invention provides an automatic compensation and calibration method for energy meter measurement errors, in order to solve the technical problems that the CORDIC algorithm is difficult to balance efficiency and accuracy, and that existing compensation methods are difficult to correct multidimensional nonlinear errors under wide temperature and wide current conditions.

[0005] An automatic compensation and calibration method for energy meter measurement error includes the following steps:

[0006] S1, synchronously acquire the voltage sampling sequence of the grid voltage and the current sampling sequence of the line current; construct the current sampling sequence into an input vector sequence, perform an improved CORDIC iterative operation on each input vector of the input vector sequence, and output the uncorrected x-axis component and the final cumulative phase angle. The improved CORDIC iterative operation includes:

[0007] Mth coarse rotation iteration: In the kth coarse rotation iteration, the rotation angle of this iteration is determined from the preset multi-level angle table based on the quadrant where the current input vector is located and the path status code that records the previous rotation direction.

[0008] P-th fine rotation iteration: In the i-th fine rotation iteration, based on the vector magnitude and rotation direction after the (i-1)-th iteration, the adjustment factor is indexed from the preset phase adjustment matrix, and the cumulative phase angle of this iteration is adjusted using the adjustment factor;

[0009] S2, based on the effective value level of the current line current and the temperature of the energy meter, select a set of polynomial coefficients from the pre-stored two-dimensional nonlinear coefficient table, perform amplitude correction on the uncorrected x-axis component, and obtain the corrected instantaneous current amplitude.

[0010] S3. The corrected instantaneous current value is reconstructed using the corrected instantaneous current amplitude and the final accumulated phase angle. The corrected instantaneous current value is multiplied by the instantaneous voltage value to obtain the instantaneous active power. The instantaneous active power is then accumulated and integrated to obtain the compensated and calibrated energy value.

[0011] Furthermore, the process of constructing the current sampling sequence into an input vector sequence includes the following steps:

[0012] The current sampling sequence was processed using a 9th-order finite impulse response filter. Perform a Hilbert transform to obtain its orthogonal component sequence. ;

[0013] Current sampling sequence Instantaneous current value x-axis component of the input vector orthogonal component sequences orthogonal components y-axis component of the input vector , constitute the input vector .

[0014] Furthermore, in S1, the specific iterative process of the coarse rotation iteration is as follows: at the beginning of the k-th coarse rotation iteration, based on the signs of the x and y components of the current input vector, it is determined whether the current input vector is in the upper half-plane or the lower half-plane of the coordinate system; then, this position information is combined with the path status code to form a composite index, wherein the path status code records the rotation direction of all iterations from the 0th to the (k-1th)th iteration.

[0015] Furthermore, the process of determining the rotation angle for this iteration from a pre-defined multi-level angle table includes the following steps:

[0016] Setting M to 4, the preset multi-level angle table contains four levels of rotation angles, namely... , , , ;

[0017] Set the path status code to a 2-bit binary number and record the rotation direction of the first two iterations;

[0018] In the k-th iteration, obtain the 2-bit code of the quadrant where the current input vector is located, concatenate the 2-bit code with the path status code to form a 4-bit index address, and look up and determine the rotation angle of this iteration in the multi-level angle table according to the index address.

[0019] Furthermore, after completing M coarse rotation iterations, fine rotation iterations are performed. In the i-th fine rotation iteration, a basic rotation is first performed according to the standard CORDIC algorithm. Based on this, an approximate vector magnitude is estimated based on the result of the previous iteration. Using this approximate vector magnitude as the row index and the rotation direction of this iteration as the column index, the corresponding fine-tuning adjustment factor is found in a two-dimensional phase adjustment matrix. The adjustment factor is then superimposed on the cumulative phase angle of this iteration to compensate for the small angular deviation caused by the change in vector magnitude.

[0020] Furthermore, the process of adjusting the cumulative phase angle of this iteration using an adjustment factor includes the following steps:

[0021] Set P to 10; set the base rotation angle for the i-th fine rotation iteration to... Where i = 1, 2, ..., 10;

[0022] The vector magnitude after the (i-1)th iteration is normalized and quantized into a 4-bit binary code as the row index; the rotation direction is used as the column index, where clockwise is 0 and counterclockwise is 1; the adjustment factor is obtained by looking up a table from a preset 16×2 phase adjustment matrix. ; the adjustment factor relative to the basic rotation angle Multiply to obtain the rotation angle of the i-th iteration. .

[0023] Furthermore, in S2, the temperature of the electricity meter measured by the temperature sensor is read, and the effective value of the line current calculated in the previous power frequency cycle is retrieved. The electricity meter temperature is divided into intervals, and the effective value of the line current is divided into levels.

[0024] Furthermore, in S2, a corresponding set of second-order polynomial coefficients is found to be... , , Substitute the uncorrected x-axis components obtained from the CORDIC iterative calculation into the correction formula. The instantaneous current amplitude after amplitude correction was calculated. ,in, This represents the uncorrected x-axis component.

[0025] Furthermore, in S3, through the formula The corrected instantaneous current value was calculated. ,in, The corrected instantaneous current amplitude at sampling time n. This is the final accumulated phase angle.

[0026] Furthermore, in S3, through the formula Instantaneous active power was calculated ,in, This represents the instantaneous voltage value of the voltage sampling sequence.

[0027] The beneficial effects are as follows: This invention improves the CORDIC algorithm by employing a two-stage iterative strategy of coarse rotation and fine rotation. The coarse rotation, based on quadrants and path status codes, can quickly approximate the target angle, accelerating the algorithm's speed. The fine rotation, combined with vector magnitude and rotation direction, can dynamically adjust the phase angle, ensuring computational accuracy. This invention establishes a two-dimensional nonlinear coefficient table indexed by the effective current value and the energy meter temperature, and selects corresponding polynomial coefficients to correct the current amplitude, compensating for nonlinear errors caused by the combined effects of load current and ambient temperature. This invention ensures that the energy meter maintains high-precision metering performance across a wide range of current and temperature variations, improving the accuracy and reliability of energy metering. Attached Figure Description

[0028] Figure 1 A flowchart of an automatic compensation calibration method for electricity meter measurement errors;

[0029] Figure 2 A two-dimensional input vector diagram of the current sampling sequence;

[0030] Figure 3This is a schematic diagram of iterative rotation. Detailed Implementation

[0031] An embodiment of the automatic compensation and calibration method for energy meter measurement error provided by the present invention:

[0032] like Figure 1 As shown, an automatic compensation calibration method for energy meter measurement error includes the following steps:

[0033] S1, synchronously acquire the voltage sampling sequence of the grid voltage and the current sampling sequence of the line current; construct the current sampling sequence into an input vector sequence, perform an improved CORDIC iterative operation on each input vector of the input vector sequence, and output the uncorrected x-axis component and the final cumulative phase angle. The improved CORDIC iterative operation includes:

[0034] Mth coarse rotation iteration: In the kth coarse rotation iteration, the rotation angle of this iteration is determined from the preset multi-level angle table based on the quadrant where the current input vector is located and the path status code that records the previous rotation direction.

[0035] P-th fine rotation iteration: In the i-th fine rotation iteration, based on the vector magnitude and rotation direction after the (i-1)-th iteration, the adjustment factor is indexed from the preset phase adjustment matrix, and the cumulative phase angle of this iteration is adjusted using the adjustment factor.

[0036] For each new instantaneous current value i[n], combined with the historical instantaneous current value i[n-64] delayed by a quarter of a power frequency cycle, a two-dimensional input vector is constructed: x0=i[n-64], y0=i[n], as follows: Figure 2 As shown, the input vector is (x0=i[n-64], y0=i[n]). This construction method is equivalent to shifting the phase of the current signal by 90° in the time domain, which facilitates the subsequent extraction of its instantaneous amplitude and phase using the CORDIC algorithm.

[0037] Multiple input vectors are generated sequentially according to the sampling time sequence to form a complete input vector sequence. The input vector sequence is then fed into the improved CORDIC arithmetic unit for processing one by one.

[0038] In one embodiment, the specific iterative process of the coarse rotation iteration is as follows: At the beginning of the k-th coarse rotation iteration (k=1, 2, ..., M), based on the signs of the x and y components of the current input vector (i.e., the vector output by the k-1th iteration), it is determined whether the current input vector is in the upper half-plane (y>0) or the lower half-plane (y<0) of the coordinate system; then, this position information is combined with the path status code to form a composite index, wherein the path status code records the rotation direction of all iterations from the 0th to the k-1th iteration.

[0039] Subsequently, the composite index is used to search the multi-level angle table stored in read-only memory to obtain the shift parameter corresponding to the preset rotation angle for this iteration. Based on this shift parameter, a large-angle rotation is performed on the current intermediate vector to quickly approach the x-axis. Simultaneously, the rotation angle is accumulated to the final accumulated phase angle to update the accumulated phase angle, and the path status code is updated. The rotation direction is recorded for use in subsequent iterations. Figure 3 As shown.

[0040] After completing M coarse rotations, the fine rotation iteration continues. The specific process is as follows: In the i-th fine rotation iteration, a basic rotation is first performed according to the standard CORDIC algorithm, with the rotation angle being the preset basic angle (i.e., arctan(2π / 3)). -i Based on this, and using the results of the previous iteration, an approximate vector magnitude is estimated. Specifically, the approximate vector magnitude can be the absolute value of the x-component. Using this approximate vector magnitude as the row index and the rotation direction of this iteration as the column index, the corresponding fine-tuning adjustment factor is found in a two-dimensional phase adjustment matrix. The adjustment factor is then superimposed on the cumulative phase angle of this iteration to compensate for the small angular deviation caused by the change in vector magnitude, such as... Figure 2 As shown, this enables higher precision phase calculation.

[0041] In one embodiment, the phase adjustment matrix is ​​shown in Table 1.

[0042] Table 1: Phase Adjustment Matrix

[0043]

[0044] After all M+P iterations, the y-axis component of the vector will approach zero, and finally the corresponding x-axis component and the final cumulative phase angle can be obtained.

[0045] In a preferred embodiment, the process of constructing the current sampling sequence into an input vector sequence includes the following steps:

[0046] The current sampling sequence was processed using a 9th-order finite impulse response (FIR) filter. Perform a Hilbert transform to obtain its orthogonal component sequence. ;

[0047] Current sampling sequence Instantaneous current value x-axis component of the input vector orthogonal component sequences orthogonal components y-axis component of the input vector , constitute the input vector .

[0048] The analog-to-digital converter outputs a series of current sampling sequences. To obtain components orthogonal to each sampling time point, the sequence is fed into a 9th-order FIR filter. The frequency response of a 9th-order FIR filter can approximate a 90-degree phase shift, i.e., a Hilbert transform. For example, when an instantaneous current value of 10.0A... When entering the processing flow, it is directly used as the x-axis component. Simultaneously, the sequence containing this sample and several samples before and after it undergoes convolution in the FIR filter, outputting a corresponding orthogonal component. For example, its value is -2.5A, which forms a two-dimensional input vector (10.0, -2.5). The magnitude of this two-dimensional input vector represents the instantaneous envelope or amplitude of the signal, and the angle between it and the x-axis represents the instantaneous phase of the signal.

[0049] In a preferred embodiment, in the k-th coarse rotation iteration, the rotation angle for this iteration is determined from a preset multi-level angle table based on the quadrant of the current input vector and the path status code recording the previous rotation direction, including the following steps:

[0050] Setting M to 4, the preset multi-level angle table contains four levels of rotation angles, namely... , , , ;

[0051] Set the path status code to a 2-bit binary number and record the rotation direction of the first two iterations;

[0052] In the k-th iteration, obtain the 2-bit code of the quadrant where the current input vector is located, concatenate the 2-bit code with the path status code to form a 4-bit index address, and look up and determine the rotation angle of this iteration in the multi-level angle table according to the index address.

[0053] In this embodiment, a coarse rotation is performed four times to quickly rotate the vector close to the X-axis. The path status code records the rotation direction of the first two rotations; for example, counter-clockwise is recorded as 1, and clockwise as 0. During the third iteration, assuming the first two rotation directions were counter-clockwise and clockwise, the path status code is 10. If the current vector is in the second quadrant, its code is 01, which is then concatenated to form a 4-bit index address 0110. Based on address 0110, the optimal rotation angle is searched in a pre-stored angle table; this angle is relative to the base angle. A fine-tuning value, which is determined by the historical path and the current quadrant, aims to compensate for the error caused by the discrete rotation angle, so that the vector can converge to the target angle more quickly.

[0054] In a preferred embodiment, in the i-th fine rotation iteration, based on the vector magnitude and rotation direction after the (i-1)-th iteration, an adjustment factor is indexed from a preset phase adjustment matrix, and the cumulative phase angle of this iteration is adjusted using the adjustment factor, including the following steps:

[0055] Set P to 10; The basic rotation angle for the secondary refinement rotation iteration is set to... Where i = 1, 2, ..., 10;

[0056] The first After the next iteration, the vector magnitude is normalized and quantized into a 4-bit binary code as the row index; the rotation direction is used as the column index, where clockwise is 0 and counterclockwise is 1; the adjustment factor is obtained by looking up a table from a preset 16×2 phase adjustment matrix. ; the adjustment factor relative to the basic rotation angle Multiply to obtain the rotation angle of the i-th iteration. .

[0057] In this embodiment, the impact of changes in the magnitude of the CORDIC algorithm on phase accuracy is compensated by adjusting a small rotation angle. In 10 fine rotation iterations, taking the 5th iteration as an example, the base rotation angle is... Calculate the vector magnitude after the 4th iteration. Assume the normalized and quantized 4-bit encoding of the vector magnitude is 1011. Simultaneously, based on the sign of the y-component of the current vector, the 5th iteration should be a counter-clockwise rotation, with a direction encoding of 1. Using row index 1011 and column index 1, find an adjustment factor from the 16×2 phase adjustment matrix. For example, 0.998. The actual rotation angle used in this iteration is... Fine adjustments to the base angle can correct accumulated phase errors.

[0058] S2, based on the effective value level of the current line current and the temperature of the energy meter, selects a set of polynomial coefficients from the pre-stored two-dimensional nonlinear coefficient table, performs amplitude correction on the uncorrected x-axis component, and obtains the corrected instantaneous current amplitude.

[0059] The temperature of the electricity meter is read from the temperature sensor, and the effective value of the line current calculated in the previous power frequency cycle is retrieved. The electricity meter temperature is divided into intervals. For example, the temperature value is divided into intervals of 20°C, and the effective value of the line current is divided into different levels such as 1%, 5%, and 10% of the rated value.

[0060] Using the aforementioned temperature and current levels as coordinates, a corresponding set of second-order polynomial coefficients is found in a pre-stored table of two-dimensional nonlinear coefficients in non-volatile memory. For example... , , Substitute the uncorrected x-axis component (i.e., the uncorrected instantaneous current amplitude) obtained from the CORDIC iterative calculation into the correction formula. The instantaneous current amplitude after amplitude correction was calculated. ,in, The x-axis component is the uncorrected component. Table 2 is an example of a two-dimensional nonlinear coefficient table.

[0061] Table 2: Table of Two-Dimensional Nonlinear Coefficients

[0062]

[0063] S3. The corrected instantaneous current value is reconstructed using the corrected instantaneous current amplitude and the final accumulated phase angle. The corrected instantaneous current value is multiplied by the instantaneous voltage value to obtain the instantaneous active power. The instantaneous active power is then accumulated and integrated to obtain the compensated and calibrated energy value.

[0064] The corrected instantaneous current amplitude obtained using S2 The final accumulated phase angle output by the CORDIC operation in S1 The corrected instantaneous current value is reconstructed through trigonometric function operations. .

[0065] The corrected instantaneous current value Multiplying the instantaneous voltage value u[n] acquired synchronously yields the instantaneous active power at that sampling moment. At each sampling moment, the instantaneous active power p[n] is accumulated into a dedicated 64-bit wide energy accumulation register. The multiplication and accumulation process is repeated at each sampling moment, and the value in the energy accumulation register continues to increase. The accumulated value is the energy metering value after full compensation and calibration.

[0066] In a preferred embodiment, the corrected instantaneous current value is reconstructed using the corrected instantaneous current amplitude and the final accumulated phase angle in S1. The instantaneous active power is obtained by multiplying the corrected instantaneous current value by the instantaneous voltage value, including:

[0067] Let the corrected instantaneous current amplitude at sampling time n be _____. The final accumulated phase angle is The instantaneous voltage value collected synchronously is u[n];

[0068] Through formula The corrected instantaneous current value was calculated. ;

[0069] Through formula Instantaneous active power was calculated .

[0070] At a certain sampling time n, assuming the instantaneous current amplitude obtained after the aforementioned correction steps is... The value is 13.5A, and the final accumulated phase angle output by the CORDIC algorithm is... The value is 0.45 rad. The instantaneous current value is calculated by reconstructing the instantaneous current value using the sine of the instantaneous current amplitude and the final accumulated phase angle. The instantaneous current value, after double correction for both instantaneous current amplitude and final accumulated phase angle, is multiplied by the instantaneous voltage value acquired at the same sampling time n. The resulting instantaneous active power... It integrates compensation for amplitude nonlinearity error and phase shift error in current measurement, and its integral result is the electrical energy, with higher accuracy than the result calculated by directly using the original sampled values.

[0071] In addition, in the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

Claims

1. An automatic compensation and calibration method for energy meter measurement error, characterized in that, Includes the following steps: S1, synchronously acquire the voltage sampling sequence of the grid voltage and the current sampling sequence of the line current; construct the current sampling sequence into an input vector sequence, perform an improved CORDIC iterative operation on each input vector of the input vector sequence, and output the uncorrected x-axis component and the final cumulative phase angle. The improved CORDIC iterative operation includes: Mth coarse rotation iteration: In the kth coarse rotation iteration, the rotation angle of this iteration is determined from the preset multi-level angle table based on the quadrant where the current input vector is located and the path status code that records the previous rotation direction. P-th fine rotation iteration: In the i-th fine rotation iteration, based on the vector magnitude and rotation direction after the (i-1)-th iteration, the adjustment factor is indexed from the preset phase adjustment matrix, and the cumulative phase angle of this iteration is adjusted using the adjustment factor; S2, based on the effective value level of the current line current and the temperature of the energy meter, select a set of polynomial coefficients from the pre-stored two-dimensional nonlinear coefficient table, perform amplitude correction on the uncorrected x-axis component, and obtain the corrected instantaneous current amplitude. S3. The corrected instantaneous current value is reconstructed using the corrected instantaneous current amplitude and the final accumulated phase angle. The corrected instantaneous current value is multiplied by the instantaneous voltage value to obtain the instantaneous active power. The instantaneous active power is then accumulated and integrated to obtain the compensated and calibrated energy value.

2. The automatic compensation calibration method for energy meter measurement error according to claim 1, characterized in that, The process of constructing an input vector sequence from a current sampling sequence includes the following steps: The current sampling sequence was processed using a 9th-order finite impulse response filter. Perform a Hilbert transform to obtain its orthogonal component sequence. ; Current sampling sequence Instantaneous current value x-axis component of the input vector orthogonal component sequences orthogonal components y-axis component of the input vector , constitute the input vector .

3. The automatic compensation calibration method for energy meter measurement error according to claim 1, characterized in that, In S1, the specific iterative process of the coarse rotation iteration is as follows: At the beginning of the kth coarse rotation iteration, based on the signs of the x and y components of the current input vector, it is determined whether the current input vector is in the upper half-plane or the lower half-plane of the coordinate system; then, this position information is combined with the path status code to form a composite index, where the path status code records the rotation direction of all iterations from the 0th to the (k-1th)th iteration.

4. The automatic compensation calibration method for energy meter measurement error according to claim 3, characterized in that, The process of determining the rotation angle for this iteration from a pre-defined multi-level angle table includes the following steps: Setting M to 4, the preset multi-level angle table contains four levels of rotation angles, namely... , , , ; Set the path status code to a 2-bit binary number and record the rotation direction of the first two iterations; In the k-th iteration, obtain the 2-bit code of the quadrant where the current input vector is located, concatenate the 2-bit code with the path status code to form a 4-bit index address, and look up and determine the rotation angle of this iteration in the multi-level angle table according to the index address.

5. The automatic compensation calibration method for energy meter measurement error according to claim 3, characterized in that, After completing M coarse rotation iterations, continue with fine rotation iterations, until the Mth iteration... In the next fine rotation iteration, a basic rotation is first performed according to the standard CORDIC algorithm. Based on this, an approximate vector magnitude is estimated based on the result of the previous iteration. Using this approximate vector magnitude as the row index and the rotation direction of this iteration as the column index, the corresponding fine-tuning adjustment factor is found in a two-dimensional phase adjustment matrix. The adjustment factor is then superimposed on the cumulative phase angle of this iteration to compensate for the small angular deviation caused by the change in vector magnitude.

6. The automatic compensation calibration method for energy meter measurement error according to claim 1, characterized in that, The process of adjusting the cumulative phase angle of this iteration using an adjustment factor includes the following steps: Set P to 10; The basic rotation angle for the secondary refinement rotation iteration is set to... Where i = 1, 2, ..., 10; The first After the next iteration, the vector magnitude is normalized and quantized into a 4-bit binary code as the row index; the rotation direction is used as the column index, where clockwise is 0 and counterclockwise is 1; the adjustment factor is obtained by looking up a table from a preset 16×2 phase adjustment matrix. ; the adjustment factor relative to the basic rotation angle Multiply to get the first... Next iteration rotation angle .

7. The automatic compensation calibration method for energy meter measurement error according to claim 1, characterized in that, In S2, the temperature of the electricity meter measured by the temperature sensor is read, and the effective value of the line current calculated in the previous power frequency cycle is retrieved. The electricity meter temperature is divided into intervals, and the effective value of the line current is divided into levels.

8. The automatic compensation calibration method for energy meter measurement error according to claim 7, characterized in that, In S2, the corresponding set of second-order polynomial coefficients is found to be: , , Substitute the uncorrected x-axis components obtained from the CORDIC iterative calculation into the correction formula. The instantaneous current amplitude after amplitude correction was calculated. ,in, This represents the uncorrected x-axis component.

9. The automatic compensation calibration method for energy metering error according to any one of claims 1-8, characterized in that, In S3, through the formula The corrected instantaneous current value was calculated. ,in, The corrected instantaneous current amplitude at sampling time n. This is the final accumulated phase angle.

10. The automatic compensation calibration method for energy meter measurement error according to claim 9, characterized in that, In S3, through the formula Instantaneous active power was calculated ,in, This represents the instantaneous voltage value of the voltage sampling sequence.

Images