Amplitude correction method and system for reducing fence effect influence during power harmonic FFT

Abstract

The invention discloses an amplitude correction method and system for reducing fence effect influence during power harmonic FFT (fast Fourier transform), and the method comprises the steps: firstly carrying out the frequency domain processing of a time domain harmonic signal, and obtaining an amplitude-frequency characteristic curve of the time domain signal; searching a frequency point location of the amplitude-frequency characteristic curve to obtain a maximum-amplitude spectral line and a secondary-amplitude spectral line adjacent to the maximum-amplitude spectral line; then calculating a module value ratio of the maximum amplitude to the adjacent secondary amplitude, constructing a function relationship between the module value ratio and the fence factor, and correcting the maximum amplitude and the adjacent secondary amplitude to obtain a correction result; and finally, based on the correction result of the maximum amplitude and the correction result of the adjacent secondary amplitude, respectively setting weight factors, and carrying out weighted correction on the real amplitude spectral line of the time domain signal. According to the method, the weights of the maximum amplitude spectral line and the second amplitude spectral line are set through the fence factors, and the corresponding correction results are given to the weights, so that the weighted correction result of the real amplitude spectral line of the measured time domain signal in the power harmonic is obtained, the accuracy of power harmonic signal measurement is effectively improved, and the influence of the fence effect is further weakened.

Description

Technical Field

[0001] The present invention belongs to the field of digital measurement of power grid signals, and particularly relates to an amplitude correction method and system for reducing the influence of the fence effect during FFT of power harmonics. Background Art

[0002] With the extensive application of new energy power generation technologies such as wind power and photovoltaic power, and the rapid development of power electronics technology, more and more non-linear loads are applied to the power system, resulting in increasingly serious harmonic pollution of the power grid signals. Power harmonics not only affect the normal operation of the power system but also deteriorate the quality of the supplied power. Therefore, timely and accurately grasping the actual situation of power harmonics existing in the power grid signals is of great significance for ensuring the safe, stable, and economic operation of the power system.

[0003] Accurate measurement of power harmonics is a prerequisite for monitoring and controlling power harmonics. One of the important links to achieve accurate measurement of power harmonics is to accurately obtain the amplitudes of power harmonic components. The existing Fast Fourier Transform (FFT) is one of the main methods for measuring and calculating power harmonics. However, when FFT is applied to the digital measurement of power harmonics, the fence effect will inevitably occur. Based on the FFT results, the accuracy of parameters such as the frequency, amplitude, and initial phase of the measured power harmonic signal will be relatively low, unable to meet the measurement requirements. Currently, to suppress the influence of the fence effect, some scholars have proposed a single-peak spectral line correction algorithm, that is, using the spectral line with the largest amplitude to correct the result obtained by FFT transformation. The verification results show that the single-peak spectral line correction algorithm can reduce the influence of the fence effect to a certain extent, but at the same time, it may lead to a large error between the corrected results of the amplitude and initial phase of the measured signal and their actual values at individual frequencies, and the maximum error may even exceed 100%. Based on the single-peak spectral line correction algorithm, some scholars have also proposed a double-peak spectral line correction algorithm. This method uses the information of the spectral line with the second-largest amplitude in addition to the information of the spectral line with the largest amplitude for amplitude correction. Specifically, it performs a weighted average on the amplitudes of the two spectral lines, and the weights used are both proportional to the amplitudes of the two spectral lines. Compared with the single-peak spectral line correction algorithm, the double-peak spectral line correction method further reduces the influence of the fence effect. However, experimental verification shows that there is still a situation where the maximum correction result error may be close to 20%, which cannot meet the actual needs of accurately measuring the amplitude of signal components. Summary of the Invention

[0004] The object of the present invention is to provide an amplitude correction method and system with higher accuracy and capable of further reducing the influence of the fence effect during FFT of power harmonics in view of the above problems existing in the prior art.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] In a first aspect, the present invention proposes an amplitude correction method for reducing the influence of the fence effect during the FFT of power harmonics, including:

[0007] S1. Perform frequency-domain processing on the time-domain harmonic signal to obtain the amplitude-frequency characteristic curve of the time-domain signal; search for the frequency points of the amplitude-frequency characteristic curve to obtain the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line adjacent to the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal;

[0008] S2. Calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, construct a functional relationship between the modulus ratio and the fence factor, and based on this functional relationship, correct the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal to obtain the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line;

[0009] S3. Based on the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line, set the weight factors for the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line, and perform weighted correction on the true amplitude spectrum line of the time-domain signal.

[0010] The S2 includes:

[0011] S21. Calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal:

[0012] When X R (p + 1) is the sub-maximum amplitude spectrum line, the modulus ratio is:

[0013]

[0014] When X R (p - 1) is the sub-maximum amplitude spectrum line, the modulus ratio is:

[0015]

[0016] In the above formula, α is the modulus ratio of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line, X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectrum line;

[0017] S22. The functional relationship constructed between the modulus ratio and the fence factor is:

[0018]

[0019] In the above formula, δ is the fence factor;

[0020] S23. Modify the amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal based on the functional relationship between the modulus ratio and the fence factor;

[0021] When X R (p + 1) is the sub-maximum amplitude spectrum line, the following formula is used for modification:

[0022]

[0023] When X R (p - 1) is the sub-maximum amplitude spectrum line, the following formula is used for modification:

[0024]

[0025] In the above formula, A1 is the modification result of the maximum amplitude spectrum line, A2 is the modification result of the sub-maximum amplitude spectrum line, and N is the total number of sampling points.

[0026] The said S3 includes:

[0027] S31. Numerically match the fence factor with the modification results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line respectively, and set the weight factors of the modification results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line according to the numerical matching results;

[0028] S32. Based on the weight factors of the modification results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line, perform weighted modification on the true amplitude spectrum line of the time-domain signal:

[0029] When X R (p + 1) is the sub-maximum amplitude spectrum line, the weighted modification result of the true amplitude spectrum line of the time-domain signal is:

[0030]

[0031] When X R (p - 1) is the sub-maximum amplitude spectrum line, the weighted modification result of the true amplitude spectrum line of the time-domain signal is:

[0032]

[0033] In the above formula, A is the weighted modification result of the true amplitude spectrum line of the time-domain signal, δ is the fence factor, X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, p is the frequency point corresponding to the maximum amplitude spectrum line, and N is the total number of sampling points.

[0034] The said S1 includes:

[0035] S11. Assume that the time-domain signal existing in the power harmonics is a single-frequency time-domain signal, and it is represented by the following formula:

[0036] x(t) = A0sin(2πf0t + θ0);

[0037] In the above formula, x(t) is a single - frequency time - domain signal, A0 is the amplitude, f0 is the frequency, and θ0 is the initial phase;

[0038] S12. Perform equally - spaced sampling on the single - frequency time - domain signal to obtain the discrete signal x(n) of this time - domain signal as:

[0039]

[0040] In the above formula, n = 0, 1, 2,..., N - 1, N is the total number of sampling points, and f s is the sampling frequency;

[0041] S13. Select a window function to truncate the discrete signal x(n) of the time - domain signal to obtain the truncated signal x R (n) as:

[0042] x R (n) = x(n)ω(n);

[0043] In the above formula, ω(n) is the time - domain discrete expression of the window function;

[0044] S14. Perform discrete Fourier transform on the truncated signal x R (n) to obtain the discrete Fourier transform X R (k) after windowing the signal as:

[0045]

[0046] In the above formula, j is the imaginary number, W(k) is the frequency - domain expression of the rectangular window function with the total number of sampling points N, k is the serial number of the spectral line corresponding to the frequency, and k0 is the serial number of the spectral line corresponding to the true frequency of the time - domain signal;

[0047] S15. Based on the discrete Fourier transform X R (k) after windowing the signal, obtain the amplitude - frequency characteristic curve |X R (k)| - k of the time - domain signal, search for the frequency points of this amplitude - frequency characteristic curve, and the maximum amplitude spectral line obtained should satisfy the following formula:

[0048] [|X R (p)| - |X R (p - 1)|]*[|X R (p + 4)| - |X R (p)|] < 0;

[0049] In the above formula, X R(p) is the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectral line; when |X R (p + 1)| > |X R (p - 1)|, take X R (p + 1) as the sub-maximum amplitude spectral line, otherwise, take X R (p - 1) as the sub-maximum amplitude spectral line.

[0050] In a second aspect, the present invention proposes an amplitude correction system for reducing the influence of the fence effect in the FFT of power harmonics, including an amplitude-frequency characteristic curve search module, an amplitude spectral line correction module, and a true amplitude spectral line weighted correction module;

[0051] The amplitude-frequency characteristic curve search module is used to perform frequency-domain processing on the time-domain harmonic signal to obtain the amplitude-frequency characteristic curve of the time-domain signal; search for the frequency points of the amplitude-frequency characteristic curve to obtain the maximum amplitude spectral line and the sub-maximum amplitude spectral line adjacent to the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal;

[0052] The amplitude spectral line correction module is used to calculate the modulus ratio of the maximum amplitude spectral line and the adjacent sub-maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal, construct a functional relationship between the modulus ratio and the fence factor, and based on this functional relationship, correct the maximum amplitude spectral line and the adjacent sub-maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal to obtain the correction results of the maximum amplitude spectral line and the adjacent sub-maximum amplitude spectral line;

[0053] The true amplitude spectral line weighted correction module is used to set the weight factors of the correction results of the maximum amplitude spectral line and the sub-maximum amplitude spectral line based on the correction results of the maximum amplitude spectral line and the adjacent sub-maximum amplitude spectral line, and perform weighted correction on the true amplitude spectral line of the time-domain signal.

[0054] The amplitude spectral line correction module includes a modulus ratio calculation unit, a functional relationship construction unit, and an amplitude spectral line correction unit;

[0055] The modulus ratio calculation unit is used to calculate the modulus ratio of the maximum amplitude spectral line and the adjacent sub-maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal:

[0056] When X R (p + 1) is the sub-maximum amplitude spectral line, the modulus ratio is:

[0057]

[0058] When X R (p - 1) is the sub-maximum amplitude spectral line, the modulus ratio is:

[0059]

[0060] In the above formula, α is the modulus ratio of the maximum amplitude spectral line to the adjacent second-largest amplitude spectral line, and X R (p) is the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectral line;

[0061] The function relationship construction unit is used to construct the following function relationship between the modulus ratio and the fence factor:

[0062]

[0063] In the above formula, δ is the fence factor;

[0064] The amplitude spectral line correction unit is used to correct the amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal based on the function relationship between the modulus ratio and the fence factor;

[0065] When X R (p + 1) is the second-largest amplitude spectral line, the following formula is used for correction:

[0066]

[0067] When X R (p - 1) is the second-largest amplitude spectral line, the following formula is used for correction:

[0068]

[0069] In the above formula, A1 is the correction result of the maximum amplitude spectral line, A2 is the correction result of the second-largest amplitude spectral line, and N is the total number of sampling points.

[0070] The true amplitude spectral line weighted correction module includes a weight factor setting unit and a weighted correction unit;

[0071] The weight factor setting unit is used to numerically match the fence factor with the correction results of the maximum amplitude spectral line and the second-largest amplitude spectral line respectively, and set the weight factors of the correction results of the maximum amplitude spectral line and the second-largest amplitude spectral line according to the numerical matching results;

[0072] The weighted correction unit is used to perform weighted correction on the true amplitude spectral line of the time-domain signal based on the weight factors of the correction results of the maximum amplitude spectral line and the second-largest amplitude spectral line:

[0073] When X R (p + 1) is the second-largest amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is:

[0074]

[0075] When X RWhen (p - 1) is the sub-large amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is:

[0076]

[0077] In the above formula, A is the weighted correction result of the true amplitude spectral line of the time-domain signal, δ is the fence factor, and X R (p) is the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal, p is the frequency point corresponding to the maximum amplitude spectral line, and N is the total number of sampling points.

[0078] The amplitude-frequency characteristic curve search module includes a time-domain signal setting unit, an equidistant sampling unit, a signal truncation unit, a Fourier transform unit, and an amplitude spectral line search unit;

[0079] The time-domain signal setting unit is used to set the time-domain signal existing in the power harmonics as a single-frequency time-domain signal, which is represented by the following formula:

[0080] x(t) = A0sin(2πf0t + θ0);

[0081] In the above formula, x(t) is a single-frequency time-domain signal, A0 is the amplitude, f0 is the frequency, and θ0 is the initial phase;

[0082] The equidistant sampling unit is used to perform equidistant sampling on the single-frequency time-domain signal, and the discrete signal x(n) of the time-domain signal is obtained as:

[0083]

[0084] In the above formula, n = 0, 1, 2,..., N - 1, N is the total number of sampling points, and f s is the sampling frequency;

[0085] The signal truncation unit is used to select a window function to truncate the discrete signal x(n) of the time-domain signal, and the truncated signal x R (n) is:

[0086] x R (n) = x(n)ω(n);

[0087] In the above formula, ω(n) is the time-domain discrete expression of the window function;

[0088] The Fourier transform unit is used to perform discrete Fourier transform on the truncated signal x R (n), and the discrete Fourier transform formula X R (k) after the signal is windowed is:

[0089]

[0090] In the above formula, j is the imaginary number, W(k) is the frequency-domain expression of the rectangular window function with the total number of sampling points being N, k is the serial number of the spectral line corresponding to the frequency, and k0 is the serial number of the spectral line corresponding to the true frequency of the time-domain signal;

[0091] The amplitude spectrum line searching unit is used to obtain the amplitude-frequency characteristic curve |X R (k)|-k of the time-domain signal based on the discrete Fourier transform formula X R (k) after windowing the signal, search for the frequency points of this amplitude-frequency characteristic curve, and the obtained maximum amplitude spectrum line should satisfy the following formula:

[0092] [|X R (p)|-|X R (p - 1)|]*[|X R (p + 1)|-|X R (p)|]<0;

[0093] In the above formula, X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, p is the frequency point corresponding to the maximum amplitude spectrum line; when |X R (p + 1)|>|X R (p - 1)|, take X R (p + 1) as the second-largest amplitude spectrum line, otherwise, take X R (p - 1) as the second-largest amplitude spectrum line.

[0094] In the third aspect, the present invention proposes an amplitude correction device for reducing the influence of the fence effect during the FFT of power harmonics, including a processor and a memory;

[0095] The memory is used to store computer program code and transmit the computer program code to the processor;

[0096] The processor is used to execute the aforementioned amplitude correction method for reducing the influence of the fence effect during the FFT of power harmonics according to the instructions in the computer program code.

[0097] In the fourth aspect, the present invention proposes a computer storage medium, on which a computer program is stored;

[0098] When the computer program is executed by a processor, it implements the steps of the aforementioned amplitude correction method for reducing the influence of the fence effect during the FFT of power harmonics.

[0099] Compared with the prior art, the beneficial effects of the present invention are:

[0100] The present invention proposes an amplitude correction method and system for reducing the influence of the fence effect in the FFT of power harmonics. The method first performs frequency-domain processing on the time-domain harmonic signal to obtain the amplitude-frequency characteristic curve of the time-domain signal; searches for the frequency points of the amplitude-frequency characteristic curve to obtain the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line adjacent to the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal; then calculates the modulus ratio between the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, constructs the functional relationship between the modulus ratio and the fence factor, and based on this functional relationship, corrects the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal to obtain the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line; finally, based on the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line, sets the weight factors for the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line, and performs weighted correction on the true amplitude spectrum line of the time-domain signal. On the one hand, the method corrects the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal through the fence factor, which can effectively improve the accuracy of the measurement of power harmonic signals; on the other hand, the method assigns weight factors to the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line respectively, and obtains the weighted correction result of the true amplitude spectrum line of the measured signal frequency, which can further weaken the influence of the fence effect, making the measured true amplitude spectrum line more accurate and more robust, and meeting the actual needs of accurately measuring the amplitude of signal components. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 Schematic diagram of the amplitude-frequency characteristic curve of two single-frequency time-domain signals described in the present invention.

[0102] Figure 2 Overall flowchart of the method described in the present invention.

[0103] Figure 3 Numerical change characteristic curve matching the numerical characteristics described in Example 1.

[0104] Figure 4 Comparison result diagram of the two methods described in Example 1.

[0105] Figure 5 Structural diagram of the system described in the present invention.

[0106] Figure 6 Structural diagram of the device described in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0107] The present invention will be further described in detail below in conjunction with the specific embodiments and the accompanying drawings.

[0108] The present invention proposes an amplitude correction method and system for reducing the influence of the fence effect during the FFT of power harmonics. Aiming at the problem of insufficient robustness in existing methods, the time-domain signals existing in the measured power harmonics are discretized, sampled, and windowed, and the discrete Fourier transform formula of the signal after windowing is derived to obtain the amplitude-frequency characteristic curve of the signal. The maximum amplitude of the discrete spectral line and the second-largest amplitude adjacent to it are searched on the curve. Subsequently, a new weight setting method is proposed to establish the functional relationship between the serial number and modulus ratio of the true spectral line of the frequency of the measured signal. Based on this functional relationship, the amplitude results of the signal obtained by directly performing the discrete Fourier transform are corrected using the frequency points corresponding to the maximum amplitude of the spectral line and the second-largest amplitude adjacent to it, respectively. This method is not only applicable to the measurement and calculation of the frequency, amplitude, and initial phase of different frequency components in power signals containing multiple high-order harmonic frequency components, but also more applicable to time-domain signals containing only a single frequency component, further reducing and weakening the influence of the fence effect on the measurement and calculation of the amplitude and initial phase of a certain frequency component of the time-domain signal during the FFT of power harmonics.

[0109] The fence effect refers to the fact that after the discrete sequence of the measured signal is transformed by FFT, the obtained discrete amplitude-frequency characteristic spectral line sequence is arranged at equal frequency intervals like a fence. Since the total number of sampling points of the measured signal for FFT transformation must satisfy the 2n relationship, for a power signal containing high-order harmonics with a fundamental frequency of 50 Hz, the frequency positions of all discrete spectral lines obtained by FFT transformation usually do not coincide with the frequency points of the actual frequency components existing in the measured signal. For example, Figure 1 As shown, two single-frequency time-domain signals with frequencies of 50 Hz and 55 Hz are established, and the amplitude-frequency characteristic curve schematic diagrams are directly obtained by performing discrete Fourier transform respectively. It can be seen from the amplitude-frequency characteristic curve schematic diagram of the time-domain signal with a frequency of 55 Hz that its amplitude calculation result is significantly affected by the fence effect.

[0110] Example 1:

[0111] As Figure 2 shown, the amplitude correction method for reducing the influence of the fence effect during the FFT of power harmonics is carried out in the following steps in sequence:

[0112] 1. Perform frequency-domain processing on the time-domain harmonic signal existing in the power harmonics to obtain the amplitude-frequency characteristic curve of the time-domain signal; search for the frequency positions of the amplitude-frequency characteristic curve to obtain the maximum amplitude spectral line and the second-largest amplitude spectral line adjacent to the maximum amplitude spectral line of the time-domain signal amplitude-frequency characteristic curve;

[0113] Let the time-domain signal existing in the power harmonics be a single-frequency time-domain signal x(t) with an amplitude of 311 V, an initial phase of 0 rad, and a frequency varying from 10 Hz to 600 Hz. Its expression is:

[0114] x(t) = A0sin(2πf0t + θ0);

[0115] In the above formula, x(t) is a single - frequency time - domain signal, A0 is the amplitude, f0 is the frequency, and θ0 is the initial phase;

[0116] Perform digital processing on the single - frequency time - domain signal. Sample the single - frequency time - domain signal at equal intervals of 0.1 s with a sampling frequency of 5000 Hz to obtain the discrete signal x(n) of the time - domain signal as follows:

[0117]

[0118] In the above formula, n = 0, 1, 2,..., N - 1, N is the total number of sampling points, f s is the sampling frequency, satisfying the Nyquist sampling theorem, that is, f s ≥ 2f0;

[0119] Select a rectangular window function to truncate the discrete signal x(n) of the time - domain signal to obtain the truncated signal x R (n) as follows:

[0120] x R (n) = x(n)ω(n);

[0121] In the above formula, ω(n) is the time - domain discrete expression of the rectangular window function;

[0122] Perform discrete Fourier transform on the truncated signal x R (n). And since the influence of the frequency components at negative frequency points on the discrete spectral line sequence in the amplitude - frequency characteristic curve is small enough, the negative - frequency components in the obtained transformation result are ignored, and the discrete Fourier transform formula X R (k) after windowing the signal is as follows:

[0123]

[0124] In the above formula, j is the imaginary number, W(k) is the frequency - domain expression of the rectangular window function with the total number of sampling points N, k is the serial number of the spectral line corresponding to the frequency, k0 is the serial number of the spectral line corresponding to the true frequency of the time - domain signal, and this serial number can be a non - integer;

[0125] Based on the discrete Fourier transform formula X R (k) after windowing the signal, draw a series of discrete spectral lines on the amplitude - frequency plane, and then draw the amplitude - frequency characteristic curve |X R (k)| - k of the time - domain signal by connecting the vertices of all spectral lines. Search all the frequency points of this amplitude - frequency characteristic curve, and the maximum - amplitude spectral line should satisfy the following formula:

[0126] [|XR (p)|-|X R (p - 1)|]*[|X R (p + 1)|-|X R (p)|]<0;

[0127] In the above formula, X R (p) is the maximum amplitude spectral line of the time - domain signal amplitude - frequency characteristic curve, p is the frequency point corresponding to the maximum amplitude spectral line; when |X R (p + 1)|>|X R (p - 1)|, take X R (p + 1) as the second - largest amplitude spectral line, otherwise, take X R (p - 1) as the second - largest amplitude spectral line.

[0128] 2. Calculate the modulus ratio of the maximum amplitude spectral line and the adjacent second - largest amplitude spectral line of the time - domain signal amplitude - frequency characteristic curve, construct the functional relationship between the modulus ratio and the fence factor, and based on this functional relationship, correct the maximum amplitude spectral line and the adjacent second - largest amplitude spectral line of the time - domain signal amplitude - frequency characteristic curve to obtain the corrected results of the maximum amplitude spectral line and the adjacent second - largest amplitude spectral line;

[0129] Calculate the modulus ratio of the maximum amplitude spectral line and the adjacent second - largest amplitude spectral line of the time - domain signal amplitude - frequency characteristic curve. The calculation rule of the modulus ratio is that the amplitude of the spectral line with a higher frequency is divided by the amplitude of the spectral line with a lower frequency:

[0130] When X R (p + 1) is the second - largest amplitude spectral line, the modulus ratio is:

[0131]

[0132] When X R (p - 1) is the second - largest amplitude spectral line, the modulus ratio is:

[0133]

[0134] In the above formula, α is the modulus ratio of the maximum amplitude spectral line and the adjacent second - largest amplitude spectral line, X R (p) is the maximum amplitude spectral line of the time - domain signal amplitude - frequency characteristic curve, p is the frequency point corresponding to the maximum amplitude spectral line;

[0135] The true spectral line number λ of the time - domain signal frequency = p + δ, where p is the integer part of the true spectral line number λ, δ is the fence factor, that is, the decimal part of the true spectral line number, and δ ∈[-1, 1]. λ and the frequency f0 satisfy the following relationship:

[0136] f0 = (p + δ)*Δf;

[0137] In the above formula, Δf is the frequency resolution;

[0138] Substitute the modulus ratios in the two cases into the discrete Fourier transform formula X R (k) after signal windowing, and construct the functional relationship between the modulus ratio and the fence factor as follows:

[0139]

[0140] Based on the functional relationship between the modulus ratio and the fence factor, let k in the discrete Fourier transform formula X R (k) after signal windowing be the frequency positions corresponding to the maximum amplitude spectral line and the frequency positions corresponding to the adjacent sub-maximum amplitude spectral lines respectively, and correct the amplitude spectral lines of the time-domain signal amplitude-frequency characteristic curve:

[0141] When X R (p + 1) is the sub-maximum amplitude spectral line, the following formula is used for correction:

[0142]

[0143]

[0144] When X R (p - 1) is the sub-maximum amplitude spectral line, the following formula is used for correction:

[0145]

[0146] In the above formula, A1 is the correction result of the maximum amplitude spectral line, A2 is the correction result of the sub-maximum amplitude spectral line, and N is the total number of sampling points.

[0147] 3. Based on the correction results of the maximum amplitude spectral line and the adjacent sub-maximum amplitude spectral line, set the weight factors of the correction results of the maximum amplitude spectral line and the sub-maximum amplitude spectral line, and perform weighted correction on the true amplitude spectral line of the time-domain signal;

[0148] Match the fence factor with the correction results of the maximum amplitude spectral line and the sub-maximum amplitude spectral line respectively in terms of numerical characteristics, and set the weight factors of the correction results of the maximum amplitude spectral line and the sub-maximum amplitude spectral line according to the numerical characteristic matching results; that is, under the condition that the sampling frequency f s and the total number of sampling points N remain unchanged, the frequency of the measured analog simulation signal starts from 10 Hz, increases gradually at an interval of 1 Hz to 600 Hz, and a series of correction results based on the maximum amplitude spectral line useX(|) and its adjacent positive or negative sub-maximum amplitude spectral line useX(| + 1) are obtained, as well as the numerical change characteristic curves of the corresponding obtained fence factors are as Figure 3 shown. By observing and matching the correction results with the numerical change characteristics of the fence factor, the weight factors are obtained;

[0149] After the weight factors are matched through numerical characteristics, the weight factors of the maximum amplitude spectral line and the second-largest amplitude spectral line based on the discrete spectral lines are taken as 1 - δ and δ respectively;

[0150] Based on the weight factors of the correction results of the maximum amplitude spectral line and the second-largest amplitude spectral line, the true amplitude spectral line of the time-domain signal is weighted and corrected:

[0151] When X R (p + 1) is the second-largest amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is:

[0152]

[0153] When X R (p - 1) is the second-largest amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is:

[0154]

[0155] In the above formula, A is the weighted correction result of the true amplitude spectral line of the time-domain signal, δ is the fence factor, X R (p) is the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal, p is the frequency point corresponding to the maximum amplitude spectral line, and N is the total number of sampling points.

[0156] To verify the effectiveness of the present invention, both the double spectral line correction method and the weighted correction method described in this scheme are used to correct the amplitude of the time-domain signal of the same frequency component. The frequency of this time-domain signal starts from 10 Hz, increases gradually at an interval of 1 Hz up to 600 Hz, and the comparison results are as Figure 4 shown.

[0157] Embodiment 2:

[0158] As Figure 5 shown, an amplitude correction system for reducing the influence of the fence effect during power harmonic FFT includes an amplitude-frequency characteristic curve search module, an amplitude spectral line correction module, and a true amplitude spectral line weighted correction module;

[0159] The amplitude-frequency characteristic curve search module is used to perform frequency-domain processing on the time-domain harmonic signal to obtain the amplitude-frequency characteristic curve of this time-domain signal; search for the frequency points of this amplitude-frequency characteristic curve to obtain the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal and the second-largest amplitude spectral line adjacent to the maximum amplitude spectral line;

[0160] The amplitude spectrum line correction module is used to calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent large amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, construct the functional relationship between the modulus ratio and the fence factor, and based on this functional relationship, correct the maximum amplitude spectrum line and the adjacent large amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal to obtain the correction results of the maximum amplitude spectrum line and the adjacent large amplitude spectrum line;

[0161] The true amplitude spectrum line weighted correction module is used to set the weight factors of the correction results of the maximum amplitude spectrum line and the second largest amplitude spectrum line based on the correction results of the maximum amplitude spectrum line and the adjacent large amplitude spectrum line, and perform weighted correction on the true amplitude spectrum line of the time-domain signal.

[0162] The amplitude-frequency characteristic curve search module includes a time-domain signal setting unit, an equally spaced sampling unit, a signal truncation unit, a Fourier transform unit, and an amplitude spectrum line search unit;

[0163] The time-domain signal setting unit is used to set the time-domain signal existing in the power harmonics as a single-frequency time-domain signal, and its expression is:

[0164] x(t) = A0sin(2πf0t + θ0);

[0165] In the above formula, x(t) is a single-frequency time-domain signal, A0 is the amplitude, f0 is the frequency, and θ0 is the initial phase;

[0166] The equally spaced sampling unit is used to perform equally spaced sampling on the single-frequency time-domain signal to obtain the discrete signal x(n) of the time-domain signal as:

[0167]

[0168] In the above formula, n = 0, 1, 2,..., N - 1, N is the total number of sampling points, and f s is the sampling frequency;

[0169] The signal truncation unit is used to select a window function to truncate the discrete signal x(n) of the time-domain signal to obtain the truncated signal x R (n) as:

[0170] x R (n) = x(n)ω(n);

[0171] In the above formula, ω(n) is the time-domain discrete expression of the window function;

[0172] The Fourier transform unit is used to perform discrete Fourier transform on the truncated signal x R (n) to obtain the discrete Fourier transform X R (k) as:

[0173]

[0174] In the above formula, j is the imaginary number, W(k) is the frequency-domain expression of the rectangular window function with a total number of sampling points N, k is the serial number of the spectral line corresponding to the frequency, and k0 is the serial number of the spectral line corresponding to the true frequency of the time-domain signal;

[0175] The amplitude spectrum line search unit is used to obtain the amplitude-frequency characteristic curve |X R (k)|-k of the time-domain signal based on the discrete Fourier transform formula X R (k) of the windowed signal, search for the frequency points of this amplitude-frequency characteristic curve, and the obtained maximum amplitude spectrum line should satisfy the following formula:

[0176] [|X R (p)|-|X R (p - 1)|]*[|X R (p + 1)|-|X R (p)|]<0;

[0177] In the above formula, X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectrum line; when |X R (p + 1)|>|X R (p - 1)|, take X R (p + 1) as the second-largest amplitude spectrum line, otherwise, take X R (p - 1) as the second-largest amplitude spectrum line.

[0178] The amplitude spectrum line correction module includes a modulus ratio calculation unit, a function relationship construction unit, and an amplitude spectrum line correction unit;

[0179] The modulus ratio calculation unit is used to calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent second-largest amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal:

[0180] When X R (p + 1) is the second-largest amplitude spectrum line, the modulus ratio is:

[0181]

[0182] When X R (p - 1) is the second-largest amplitude spectrum line, the modulus ratio is:

[0183]

[0184] In the above formula, α is the modulus ratio of the maximum amplitude spectrum line and the adjacent second-largest amplitude spectrum line, X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectrum line;

[0185] The function relationship construction unit is used to construct the function relationship between the modulus ratio and the fence factor as follows:

[0186]

[0187] In the above formula, δ is the fence factor;

[0188] The amplitude spectrum line correction unit is used to correct the amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal based on the function relationship between the modulus ratio and the fence factor;

[0189] When X R (p + 1) is the sub-maximum amplitude spectrum line, the following formula is used for correction:

[0190]

[0191] When X R (p - 1) is the sub-maximum amplitude spectrum line, the following formula is used for correction:

[0192]

[0193] In the above formula, A1 is the correction result of the maximum amplitude spectrum line, A2 is the correction result of the sub-maximum amplitude spectrum line, and N is the total number of sampling points.

[0194] The true amplitude spectrum line weighted correction module includes a weight factor setting unit and a weighted correction unit;

[0195] The weight factor setting unit is used to numerically match the fence factor with the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line respectively, and set the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line according to the numerical matching results;

[0196] The weighted correction unit is used to perform weighted correction on the true amplitude spectrum line of the time-domain signal based on the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line:

[0197] When X R (p + 1) is the sub-maximum amplitude spectrum line, the weighted correction result of the true amplitude spectrum line of the time-domain signal is:

[0198]

[0199] When X R (p - 1) is the sub-maximum amplitude spectrum line, the weighted correction result of the true amplitude spectrum line of the time-domain signal is:

[0200]

[0201] In the above formula, A is the weighted correction result of the true amplitude spectrum line of the time-domain signal, δ is the fence factor, and X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, p is the frequency point corresponding to the maximum amplitude spectrum line, and N is the total number of sampling points.

[0202] Embodiment 3:

[0203] As Figure 6 shown, an amplitude correction device for reducing the influence of the fence effect in the power harmonic FFT includes a processor and a memory;

[0204] The memory is used to store computer program code and transmit the computer program code to the processor;

[0205] The processor is used to execute the amplitude correction method for reducing the influence of the fence effect in the power harmonic FFT according to the instructions in the computer program code in Embodiment 1.

[0206] Embodiment 4:

[0207] A computer storage medium, on which a computer program is stored;

[0208] When the computer program is executed by a processor, the steps of the amplitude correction method for reducing the influence of the fence effect in the power harmonic FFT described in this solution are implemented.

Claims

1. An amplitude correction method for reducing the influence of the fence effect in power harmonic FFT, characterized in that: The method includes: S1. Perform frequency-domain processing on the time-domain harmonic signal to obtain the amplitude-frequency characteristic curve of the time-domain signal; search for the frequency points of the amplitude-frequency characteristic curve to obtain the maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve and the sub-maximum amplitude spectrum line adjacent to the maximum amplitude spectrum line; S2. Calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve, construct the functional relationship between the modulus ratio and the fence factor, and based on this functional relationship, correct the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve to obtain the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line; S3. Based on the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line, set the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line, and perform weighted correction on the true amplitude spectrum line of the time-domain signal.

2. The amplitude correction method for reducing the influence of the fence effect in power harmonic FFT according to claim 1, characterized in that: The S2 includes: S21. Calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve: When X R (p + 1) is the second largest amplitude spectral line, the modulus ratio is: When X R When (p - 1) is the second largest amplitude spectral line, the modulus ratio is: In the above formula, α is the modulus ratio of the maximum amplitude spectrum line to the adjacent second-largest amplitude spectrum line, and X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectrum line; S22. The functional relationship between the modulus ratio and the fence factor is constructed as: In the above formula, δ is the fence factor; S23. Based on the functional relationship between the modulus ratio and the fence factor, correct the amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve; When X R When (p + 1) is the second largest amplitude spectral line, the following formula is used for correction: When X R When (p - 1) is the second largest amplitude spectral line, the following formula is used for correction: In the above formula, A1 is the correction result of the maximum amplitude spectrum line, A2 is the correction result of the sub-maximum amplitude spectrum line, and N is the total number of sampling points.

3. The amplitude correction method for reducing the influence of the fence effect in power harmonic FFT according to claim 1, characterized in that: The S3 includes: S31. Numerically match the fence factor with the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line respectively, and set the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line according to the numerical matching results; S32. Based on the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line, perform weighted correction on the true amplitude spectrum line of the time-domain signal: When X R When (p + 1) is the next largest amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is as follows: When X R When (p - 1) is the second largest amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is as follows: In the above formula, A is the weighted correction result of the true amplitude spectrum line of the time-domain signal, δ is the fence factor, and X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, p is the frequency point corresponding to the maximum amplitude spectrum line, and N is the total number of sampling points.

4. The amplitude correction method for reducing the influence of the fence effect in power harmonic FFT according to claim 1, characterized in that: The S1 includes: S11. Assume that the time-domain signal existing in the power harmonic is a single-frequency time-domain signal, and is represented by the following formula: x(t) = A0sin(2πf0t + θ0); In the above formula, x(t) is a single-frequency time-domain signal, A0 is the amplitude, f0 is the frequency, and θ0 is the initial phase; S12. Perform equally spaced sampling on the single-frequency time-domain signal to obtain the discrete signal x(n) of the time-domain signal as: In the above formula, n = 0, 1, 2, ..., N - 1, where N is the total number of sampling points and f s is the sampling frequency; S13. Select a window function to truncate the discrete signal x(n) of the time-domain signal to obtain a truncated signal x R (n) as follows: x R (n) = x(n)ω(n); In the above formula, ω(n) is the time-domain discrete expression of the window function; S14. Perform a discrete Fourier transform on the truncated signal x R (n) to obtain the discrete Fourier transform expression X R (k) after windowing the signal as follows: In the above formula, j is an imaginary number, W(k) is the frequency-domain expression of the rectangular window function with the total number of sampling points N, k is the serial number of the frequency corresponding spectrum line, and k0 is the serial number of the spectrum line corresponding to the true frequency of the time-domain signal; S15. Based on the discrete Fourier transform formula X of the signal after windowing, R (k), obtain the amplitude-frequency characteristic curve |X of the time-domain signal R (k)|-k, search for the frequency points of this amplitude-frequency characteristic curve, and the maximum amplitude spectral line obtained should satisfy the following formula: [|X R (p)|-|X R (p - 1)|]*[|X R (p + 1)|-|X R (p)|]>0; In the above formula, X R (p) is the maximum amplitude spectral line of the time-domain signal amplitude-frequency characteristic curve, and p is the frequency point corresponding to the maximum amplitude spectral line; when |X R (p + 1)| > |X R (p - 1)|, take X R (p + 1) as the second-largest amplitude spectral line; otherwise, take X R (p - 1) as the second-largest amplitude spectral line.

5. An amplitude correction system for reducing the influence of the fence effect during the FFT of power harmonics, characterized in that the system includes an amplitude-frequency characteristic curve search module, an amplitude spectrum line correction module, and a true amplitude spectrum line weighted correction module; the amplitude-frequency characteristic curve search module is used to perform frequency-domain processing on the time-domain harmonic signal to obtain the amplitude-frequency characteristic curve of the time-domain signal; search for the frequency points of the amplitude-frequency characteristic curve to obtain the maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve and the sub-maximum amplitude spectrum line adjacent to the maximum amplitude spectrum line; the amplitude spectrum line correction module is used to calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve, construct the functional relationship between the modulus ratio and the fence factor, and based on this functional relationship, correct the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve to obtain the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line; the true amplitude spectrum line weighted correction module is used to set the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line based on the correction results of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line, and perform weighted correction on the true amplitude spectrum line of the time-domain signal.

6. The amplitude correction system for reducing the influence of the fence effect during the FFT of power harmonics according to claim 5, characterized in that the amplitude spectrum line correction module includes a modulus ratio calculation unit, a functional relationship construction unit, and an amplitude spectrum line correction unit; the modulus ratio calculation unit is used to calculate the modulus ratio of the maximum amplitude spectrum line and the adjacent sub-maximum amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve: When X R When (p + 1) is the second largest amplitude spectral line, the modulus ratio is: When X R (p - 1) is the second largest amplitude spectral line, the modulus ratio is: In the above formula, α is the modulus ratio of the maximum amplitude spectral line to the adjacent sub-maximum amplitude spectral line, and X R (p) is the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectral line; the functional relationship construction unit is used to construct the following functional relationship between the modulus ratio and the fence factor: In the above formula, δ is the fence factor; the amplitude spectrum line correction unit is used to correct the amplitude spectrum line of the time-domain signal amplitude-frequency characteristic curve based on the functional relationship between the modulus ratio and the fence factor; When X R When (p + 1) is the second largest amplitude spectral line, the following formula is used for correction: When X R When (p - 1) is the second largest amplitude spectral line, the following formula is used for correction: In the above formula, A1 is the correction result of the maximum amplitude spectrum line, A2 is the correction result of the sub-maximum amplitude spectrum line, and N is the total number of sampling points.

7. The amplitude correction system for reducing the influence of the fence effect during the FFT of power harmonics according to claim 5, characterized in that the true amplitude spectrum line weighted correction module includes a weight factor setting unit and a weighted correction unit; the weight factor setting unit is used to numerically match the fence factor with the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line respectively, and set the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line according to the numerical matching results; the weighted correction unit is used to perform weighted correction on the true amplitude spectrum line of the time-domain signal based on the weight factors of the correction results of the maximum amplitude spectrum line and the sub-maximum amplitude spectrum line: When X R When (p + 1) is the second largest amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is as follows: When X R When (p - 1) is the sub-large amplitude spectral line, the weighted correction result of the true amplitude spectral line of the time-domain signal is as follows: In the above formula, A is the weighted correction result of the true amplitude spectrum line of the time-domain signal, δ is the fence factor, and X R (p) is the maximum amplitude spectrum line of the amplitude-frequency characteristic curve of the time-domain signal, p is the frequency point corresponding to the maximum amplitude spectrum line, and N is the total number of sampling points.

8. The amplitude correction system for reducing the influence of the fence effect during the FFT of power harmonics according to claim 5, characterized in that the amplitude-frequency characteristic curve search module includes a time-domain signal setting unit, an equidistant sampling unit, a signal truncation unit, a Fourier transform unit, and an amplitude spectrum line search unit; the time-domain signal setting unit is used to set the time-domain signal existing in the power harmonics as a single-frequency time-domain signal, and is represented by the following formula: x(t) = A0sin(2vf0t + θ0); In the above formula, x(t) is a single-frequency time-domain signal, A0 is the amplitude, f0 is the frequency, and θ0 is the initial phase; The equally-spaced sampling unit is used to perform equally-spaced sampling on the single-frequency time-domain signal, and the discrete signal x(n) of the time-domain signal is obtained as: In the above formula, n = 0, 1, 2, ..., N - 1, where N is the total number of sampling points, and f s is the sampling frequency; The signal truncation unit is used to select a window function to truncate the discrete signal x(n) of the time-domain signal, and obtain the truncated signal x R (n) as follows: x R (n) = x(n)ω(n); In the above formula, ω(n) is the time-domain discrete expression of the window function; The Fourier transform unit is used to perform a discrete Fourier transform on the truncated signal x R (n) to obtain the discrete Fourier transform X of the windowed signal R (k) as follows: In the above formula, j is the imaginary unit, W(k) is the frequency-domain expression of the rectangular window function with a total number of sampling points N, k is the serial number of the spectral line corresponding to the frequency, and k0 is the serial number of the spectral line corresponding to the true frequency of the time-domain signal; The amplitude spectrum line search unit is used to obtain the amplitude-frequency characteristic curve |X R (k)|-k of the time-domain signal based on the discrete Fourier transform X R (k) after windowing the signal, search for the frequency points of the amplitude-frequency characteristic curve, and the obtained maximum amplitude spectrum line should satisfy the following formula: [|X R (p)|-|X R (p - 1)|]*[|X R (p + 1)|-|X R (p)|]<0; In the above formula, X R (p) is the maximum amplitude spectral line of the amplitude-frequency characteristic curve of the time-domain signal, and p is the frequency point corresponding to the maximum amplitude spectral line; when |X R (p + 1)| > |X R (p - 1)|, take X R (p + 1) as the second-largest amplitude spectral line, otherwise, take X R (p - 1) as the second-largest amplitude spectral line.

9. An amplitude correction device for reducing the influence of the fence effect in power harmonic FFT, characterized in that it includes a processor and a memory; The memory is used to store the computer program code and transmit the computer program code to the processor; The processor is used to execute the amplitude correction method for reducing the influence of the fence effect in power harmonic FFT according to the instructions in the computer program code in any one of claims 1-4.

10. A computer storage medium, on which a computer program is stored, characterized in that: When the computer program is executed by a processor, the steps of the amplitude correction method for reducing the influence of the fence effect in power harmonic FFT according to any one of claims 1-4 are realized.

Images