Method for detecting frequency and phase of voltage or current signals in a power system

Abstract

This invention discloses a method for detecting the frequency and phase of voltage or current signals in a power system, relating to the field of signal detection technology. The specific steps are as follows: acquiring voltage or current signals from the power system using sensors; windowing the voltage or current signals to obtain windowed voltage or current signals; performing spectral analysis on the windowed voltage or current signals to obtain their frequency distribution; obtaining the phase of the voltage or current signals and initial parameters of a neural network based on the frequency distribution; training the neural network model to obtain a frequency calculation model; and obtaining the frequency of the voltage or current signals based on the phase and frequency calculation model. This invention not only improves the accuracy of measurement results but also enables rapid acquisition of the phase and frequency of the power system, which is beneficial for power system control and stability.

Description

Technical Field

[0001] This invention relates to the field of signal detection technology, and more specifically to a method for detecting the frequency and phase of voltage or current signals in a power system. Background Technology

[0002] Frequency measurement, harmonic measurement, and power measurement in power systems are essentially measurements of sinusoidal parameters. Fourier transform and other methods are fundamental for sinusoidal parameter measurement and have wide applications in power systems. However, with the development of sinusoidal measurement technology, the problems with Fourier transform have become increasingly prominent, such as weak anti-interference capability and insufficient detection accuracy, making it difficult to further meet the requirements for high-precision calculation of sinusoidal parameters.

[0003] Therefore, how to improve the sine measurement method to increase measurement accuracy is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] In view of this, the present invention provides a method for detecting the frequency and phase of voltage or current signals in a power system, thereby solving the above-mentioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for detecting the frequency and phase of voltage or current signals in a power system, comprising the following steps:

[0007] Using sensors to collect voltage or current signals from the power system;

[0008] The voltage or current signal is windowed using a four-term third-order Nuttall window function to obtain the windowed voltage or current signal y(n);

[0009] The frequency distribution of a voltage or current signal y(n) is obtained by performing spectral analysis on the voltage or current signal y(n) using the frequency domain convolution theorem and Fourier transform properties.

[0010] The phase of the voltage or current signal and the initial parameters of the neural network are obtained based on the frequency distribution of the voltage or current signal, and the neural network model is trained to obtain the frequency calculation model.

[0011] The frequency of a voltage or current signal is obtained based on a calculation model that calculates the phase and frequency of the voltage or current signal.

[0012] Optionally, the windowing process can be completed as follows:

[0013] Preprocess the voltage or current signal to obtain a voltage or current signal sequence;

[0014] Digital filtering is performed on a voltage or current signal sequence to obtain a filtered voltage or current signal sequence.

[0015] The filtered voltage or current signal sequence is weighted using a convolution window, and the data with an interval of N are superimposed pairwise to obtain the windowed voltage or current signal y(n).

[0016] Optionally, the expression for the windowed voltage or current signal y(n) is:

[0017] y(n)=[ω(n)x(n)+ω(nN)x(nN)];

[0018] In the formula, ω(n) is a convolution window function of length N; x(n) is the original voltage signal or current signal sequence; ω(nN) is a convolution window function of length nN; and x(nN) is the superimposed voltage signal or current signal sequence.

[0019] Optionally, the expression for the frequency distribution of the voltage or current signal is:

[0020]

[0021] In the formula, F g For window spectrum; ω k ω0 is the angular frequency of the k-th spectral line; ω0 is the initial angular frequency; θ0 is the initial phase.

[0022] Optionally, the phase of the voltage or current signal is:

[0023]

[0024] In the formula, A is the amplitude; i is the number of spectral lines.

[0025] Optionally, the steps for obtaining the initial parameters of the neural network are as follows: obtain the parameter thresholds and iteration conditions of the neural network based on the frequency distribution of the voltage or current signal.

[0026] Its advantage lies in the fact that by adding a neural network, the accuracy of the frequency values ​​obtained can be increased while increasing the frequency calculation speed.

[0027] Optionally, after setting the initial parameters of the God network, the windowed voltage or current signal sequence can be used as training samples.

[0028] Optionally, the expression for the frequency of the voltage or current signal is:

[0029]

[0030] In the formula, Y ijA represents the elements in the output sample; A is the amplitude. t0 represents the phase of the voltage or current signal; t0 represents the sampling time.

[0031] As can be seen from the above technical solution, compared with the prior art, the present invention provides a method for detecting the frequency and phase of voltage or current signals in a power system. By combining full-phase FFT and neural networks, it not only improves the accuracy of measurement results, but also can quickly obtain the phase and frequency of the power system, which is beneficial to the control and stability of the power system. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0033] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] This invention discloses a method for detecting the frequency and phase of voltage or current signals in a power system, such as... Figure 1 As shown, the specific steps are as follows:

[0036] Step 1: Use sensors to collect voltage or current signals from the power system;

[0037] Step 2: Apply a four-term third-order Nuttall window function to the voltage or current signal to obtain the windowed voltage signal y(n); the specific steps are as follows:

[0038] Step 21: Preprocess the voltage or current signal to obtain a voltage or current signal sequence, which includes:

[0039] The data from 2N-1 sampling points are sampled sequentially according to the sampling time, and the sequence length is determined according to the sampling frequency, thereby forming the original voltage signal or current signal sequence.

[0040] The original voltage or current signal sequence is periodically extended, then circularly shifted, and the center sample points are aligned to form a new sequence, which is the final voltage or current signal sequence.

[0041] Step 22: Perform digital filtering on the voltage signal or current signal sequence to obtain the filtered voltage signal or current signal sequence;

[0042] In this embodiment, the digital filtering uses an arithmetic mean filtering algorithm, which adds up multiple continuous discrete values ​​and outputs the arithmetic mean as the final value, thereby forming a filtered voltage signal or current signal sequence.

[0043] Step 23: Use a convolution window to weight the filtered voltage or current signal sequence, and superimpose the data with an interval of N in pairs to obtain the windowed voltage or current signal y(n).

[0044] A convolution window of length 2N-1 is used to weight the (2N-1) data points before and after the center sample point; then, the data points with an interval of N are superimposed pairwise to form N data points; the windowed voltage or current signal y(n) is obtained, and the expression for y(n) is:

[0045] y(n)=[ω(n)x(n)+ω(nN)x(nN)];

[0046] In the formula, ω(n) is a convolution window function of length N; x(n) is a voltage signal or current signal sequence; ω(nN) is a convolution window function of length nN; and x(nN) is the superimposed voltage signal or current signal sequence.

[0047] Wherein, the convolution window ω is the convolution of the front window ω1 of length N and the flipped back window ω2, that is:

[0048] ω(n)=ω1(n)*ω2(-n), n∈(-N+1,N-1);

[0049] The window spectrum expression is:

[0050] W(jω)=|Fg(ω)| 2 .

[0051] Step 3: Perform spectral analysis on the voltage signal y(n) using the frequency domain convolution theorem and Fourier transform properties to obtain the frequency distribution of the voltage or current signal, which is expressed as:

[0052]

[0053] In the formula, F g For window spectrum; ω kω0 is the angular frequency of the k-th spectral line; ω0 is the initial angular frequency; θ0 is the initial phase.

[0054] Step 4: Based on the frequency distribution of the voltage or current signal, obtain the phase of the voltage or current signal and the initial parameters of the neural network, train the neural network model, and obtain the frequency calculation model.

[0055] The steps for obtaining the phase of a voltage or current signal are as follows: determine the peak spectral line in the full-phase spectrum, then use bi-line interpolation FFT to select the maximum spectral lines on both sides of the peak spectral line, and use the introduced auxiliary parameter α = k0 - k1 - 0.5 to calculate the phase. The calculation formula is as follows:

[0056]

[0057] In the formula, A is the amplitude; i is the number of spectral lines.

[0058] In this embodiment, the frequency band distribution range is determined based on the frequency distribution, thereby determining the initial parameters of the neural network, such as the maximum number of iterations, the default learning rate, the momentum factor, and the error threshold. The windowed voltage or current signal sequence is used as a training sample to optimize the weights so that their error values ​​reach the standard, thus obtaining the frequency calculation model.

[0059] In this embodiment, an ANN network is selected as the neural network.

[0060] Step 5: Obtain the frequency of the voltage or current signal based on the phase and frequency calculation model of the voltage or current signal; the expression for the frequency of the voltage or current signal is:

[0061]

[0062] In the formula, Y ij Let A be the element in the i-th row and j-th column of the output sample; A is the amplitude. t0 represents the phase of the voltage or current signal; t0 represents the sampling time.

[0063] In this embodiment, the combination of full-phase FFT and neural network not only improves the accuracy of measurement results, but also enables the rapid acquisition of the phase and frequency of the power system, which is beneficial to the control and stability of the power system.

[0064] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0065] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of detecting the frequency and phase of a voltage or current signal in a power system, characterized by, The specific steps are as follows: Using sensors to collect voltage or current signals from the power system; The voltage or current signal is windowed using a four-term third-order Nuttall window function to obtain the windowed voltage or current signal y(n); The frequency distribution of a voltage or current signal y(n) is obtained by performing spectral analysis on the voltage or current signal y(n) using the frequency domain convolution theorem and Fourier transform properties. The phase of the voltage or current signal and the initial parameters of the neural network are obtained based on the frequency distribution of the voltage or current signal, and the neural network model is trained to obtain the frequency calculation model. The frequency of a voltage or current signal is obtained based on a calculation model that calculates the phase and frequency of the voltage or current signal.

2. The method for detecting the frequency and phase of a voltage or current signal in a power system according to claim 1, characterized in that, The steps for adding windows are as follows: Preprocess the voltage or current signal to obtain a voltage or current signal sequence; Digital filtering is performed on a voltage or current signal sequence to obtain a filtered voltage or current signal sequence. The filtered voltage or current signal sequence is weighted using a convolution window, and the data with an interval of N are superimposed pairwise to obtain the windowed voltage or current signal y(n).

3. The method for detecting the frequency and phase of a voltage or current signal in a power system according to claim 2, characterized in that, The expression for the windowed voltage or current signal y(n) is: ； In the formula, Let N be a convolution window function of length N; It is the original voltage or current signal sequence; Let be a convolution window function of length nN; It is the superimposed sequence of voltage or current signals.

4. The method for detecting the frequency and phase of a voltage or current signal in a power system according to claim 1, characterized in that, The expression for the frequency distribution of a voltage or current signal is: ； In the formula, For window spectrum; Let be the angular frequency of the k-th spectral line; The initial angular frequency; Let A be the initial phase and A be the amplitude.

5. The method for detecting the frequency and phase of a voltage or current signal in a power system according to claim 1, characterized in that, The phase of a voltage or current signal is: ； In the formula, A is the amplitude; i is the number of spectral lines; These are auxiliary parameters.

6. The method for detecting the frequency and phase of a voltage or current signal in a power system according to claim 1, characterized in that, The steps for obtaining the initial parameters of a neural network are as follows: obtain the parameter thresholds and iteration conditions of the neural network based on the frequency distribution of the voltage or current signal.

7. The method for detecting the frequency and phase of a voltage or current signal in a power system according to claim 1, characterized in that, After setting the initial parameters of the God Network, the windowed voltage or current signal sequence is used as the training sample.

8. The method for detecting the frequency and phase of a voltage or current signal in a power system according to claim 1, characterized in that, The expression for the frequency of a voltage or current signal is: ； In the formula, A represents the elements in the output sample; A is the amplitude. t0 represents the phase of the voltage or current signal; t0 represents the sampling time.

Images