Low-frequency oscillation identification method and low-frequency oscillation identification system of power system

Abstract

The invention relates to the technical field of damping control of an electrochemical energy storage power station, and provides a low-frequency oscillation identification method and a low-frequency oscillation identification system of a power system.The method comprises the steps that 1, an original signal of the power system is sampled according to the sampling rate with the sampling point number being N in a power frequency period, and low-pass filtering is conducted on the sampled signal to obtain a low-frequency signal; 2) carrying out sine Fourier calculation at a fixed frequency of 1 Hz on the low-frequency signal to obtain phasors at three moments with a * N / m difference; 3) calculating the real-time frequency of the power system by adopting a phasor frequency measurement method according to the real parts and imaginary parts of the phasors at the three moments; and 4) calculating a real-time frequency average value, and when the duration of the real-time frequency average value in the low-frequency range exceeds a certain time rated value, considering that low-frequency oscillation occurs. According to the method, the frequency in the range of 0.2 Hz to 2.5 Hz is calculated, and the low-frequency oscillation detection of the system is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of damping control of electrochemical energy storage power stations, and mainly to a low-frequency oscillation identification method and a low-frequency oscillation identification system of a power system. Background Art

[0002] In recent years, with the continuous development of power systems, the scale of electrochemical energy storage power stations has continued to expand, and research on maintaining the stable operation of electrochemical energy storage stations has also developed rapidly. To ensure the safe operation of energy storage stations, when the electrochemical energy storage system experiences low-frequency oscillations of 0.2Hz to 2.5Hz, and the frequency oscillation amplitude at the grid connection point is greater than 0.003Hz, the electrochemical energy storage station needs to adjust the active power through additional damping control to suppress the low-frequency power. To achieve low-frequency oscillation suppression, it is necessary to identify low-frequency oscillations of 0.2Hz to 2.5Hz.

[0003] In traditional power systems, frequency measurement methods often identify frequencies within the range of 50Hz±5Hz. For example, Chinese invention application publication number CN109142863A discloses a method and system for measuring power system frequency. This invention uses a low-frequency signal to perform a sinusoidal Fourier transform calculation to obtain a phasor value, and then uses the phasor frequency measurement method to calculate the real-time frequency of the power system. However, the sinusoidal Fourier transform algorithm only has good low-pass filtering characteristics within the range of 50-100Hz. Due to the limitations of the sinusoidal Fourier transform algorithm, this method cannot identify low-frequency oscillations between 0.2Hz and 2.5Hz. Summary of the Invention

[0004] The purpose of the present invention is to provide a low-frequency oscillation identification method and a low-frequency oscillation identification system for a power system, so as to solve the problem that the frequency calculation method in the existing power system cannot identify low-frequency oscillations of 0.2Hz to 2.5Hz.

[0005] In order to solve the above technical problems, the present invention provides a low-frequency oscillation identification method, which comprises the following steps:

[0006] 1) The original signal of the power system is sampled at a sampling rate of N sampling points per power frequency cycle, and the sampled signal is low-pass filtered to obtain a low-frequency signal;

[0007] 2) Perform a sinusoidal Fourier transform calculation on the low-frequency signal at a fixed frequency of 1 Hz to obtain the phase quantities at three times with a phase difference of a*N / m power frequency cycles, where 20 ≥ a ≥ 5, m ≥ 3, and m is divisible by N;

[0008] 3) Calculate the real-time frequency of the power system using the phasor frequency measurement method based on the real and imaginary parts of the phasors at three moments;

[0009] 4) Calculate the real-time frequency average value. When the duration of the real-time frequency average value in the low-frequency range exceeds a certain rated time, it is considered that low-frequency oscillation occurs.

[0010] Furthermore, a sinusoidal Fourier transform with a fixed frequency of 1 Hz is performed on the low-frequency signal, and the expressions of the real and imaginary parts of the phasor are obtained as follows:

[0011]

[0012] Among them, i is the sampling number, u(i) is the low-frequency signal, U re is the real part of the phasor, U im is the imaginary part of the phasor, N1=50N.

[0013] Furthermore, the real-time frequency of the power system is expressed as:

[0014]

[0015] Where p = a*N / m, T s is the sampling interval, U re (i) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i, U im (i) is the imaginary part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i, U re (ip) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time ip, U im (ip) is the imaginary part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time ip, U re (i-2p) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i-2p, U im (i-2p) is the imaginary part of the phasor obtained by sinusoidal Fourier calculation of the low-frequency signal of the power system at time i-2p.

[0016] Furthermore, the real-time frequency average value is calculated by adding the real-time frequencies corresponding to the sampling points within one power frequency cycle and dividing the sum by the number of sampling points to obtain the real-time frequency average value.

[0017] Furthermore, the low-pass filtering method is to filter the sampled power system sampling signal through a Chebyshev filter.

[0018] Furthermore, the low frequency range is 0.198 Hz to 2.502 Hz.

[0019] Furthermore, the time rating is 10 ms.

[0020] A low-frequency oscillation identification system for an electric power system comprises a processor, wherein the processor adopts the low-frequency oscillation identification method to perform low-frequency oscillation identification.

[0021] The beneficial effects of the present invention are as follows: As an improved invention, the present invention performs a sinusoidal Fourier transform calculation at a fixed frequency of 1Hz on the low-frequency signal, eliminating the influence of the power frequency, and obtains phasors at three moments with a phase difference of 10*N / m power frequency cycle. Subsequently, the real-time frequency of the power system is calculated using the phasor frequency measurement method. The tolerance of the sinusoidal Fourier transform calculation near the fundamental frequency (i.e., 1Hz) is used to accurately calculate the frequency within the range of 0.2Hz to 2.5Hz. At the same time, the duration that the real-time frequency average value is within the low-frequency range is used to accurately determine whether the system is in a low-frequency oscillation state, thereby realizing the identification of low-frequency oscillation in the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic flow chart of the method of the present invention;

[0023] Figure 2 is the frequency value calculated when the low-frequency oscillation frequency is 0.2Hz;

[0024] Figure 3 is the frequency value calculated when the low-frequency oscillation frequency is 0.5Hz;

[0025] Figure 4 is the frequency value calculated when the low-frequency oscillation frequency is 0.75Hz;

[0026] Figure 5 is the frequency value calculated when the low-frequency oscillation frequency is 1.05Hz;

[0027] Figure 6 is the frequency value calculated when the low-frequency oscillation frequency is 1.25Hz;

[0028] Figure 7 is the frequency value calculated when the low-frequency oscillation frequency is 1.5Hz;

[0029] Figure 8 is the frequency value calculated when the low-frequency oscillation frequency is 1.75Hz;

[0030] Figure 9 is the frequency value calculated when the low-frequency oscillation frequency is 2.05Hz;

[0031] Figure 10 It is the frequency value calculated when the low-frequency oscillation frequency is 2.5Hz. DETAILED DESCRIPTION

[0032] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0033] The present invention performs a sinusoidal Fourier calculation with a fixed frequency of 1 Hz on the low-frequency signal. The sinusoidal Fourier calculation with a frequency of 1 Hz can eliminate the influence of the power frequency, thereby accurately calculating the low-frequency frequency of the power system in the range of 0.2 Hz to 2.5H, thereby realizing low-frequency oscillation identification.

[0034] Implementation method of low-frequency oscillation identification method

[0035] The present invention proposes a low frequency oscillation identification method, such as Figure 1 As shown, the specific steps are:

[0036] 1) Sample and preprocess the original signal of the power system.

[0037] First, the power system raw signal is periodically sampled, sampling the power system raw signal N times within one power frequency cycle to obtain the power system sampled signal U0(i). In practice, the utilization rate of electrochemical energy storage power station devices is generally 24 or 32. In the present invention, N ≥ 12. Subsequently, the power system sampled signal U0(i) is low-pass filtered using a Chebyshev filter to obtain the power system low-frequency signal U(i).

[0038] 2) Perform sinusoidal Fourier transform calculation on the low-frequency signal with a fixed frequency of 1 Hz.

[0039] The low-frequency signal U(i) is subjected to a sinusoidal Fourier calculation with a fixed frequency of 1 Hz to obtain the real and imaginary parts of the phasors at three moments with a phase difference of a*N / m, where 20 ≥ a ≥ 5, m ≥ 3, and m should be divisible by N to ensure that a*N / m is an integer. As an embodiment, a=10 is taken.

[0040] The expression formula for the real and imaginary parts of the phasor at three moments with a phase difference of 10*N / m is:

[0041]

[0042] Among them, i is the sampling number, U re (i) is the real part of the phasor obtained by Fourier transform calculation of the fixed 1Hz frequency sinusoidal low-frequency signal of the power system, U im (i) is the imaginary part of the phasor obtained by Fourier transform calculation of the fixed 1Hz frequency sine of the power system low-frequency signal, where N1 = 50 / 1 × N = 50N. To facilitate calculation, the cos(2*i*π / N1) and sin(2*i*π / N1) corresponding to each sampling number i can be calculated in advance to form a constant list. In actual use, the corresponding values ​​of cos(2*i*π / N1) and sin(2*i*π / N1) can be directly obtained from the constant list according to the sampling number i.

[0043] 3) Calculate the real-time frequency of the power system.

[0044] Based on the obtained phasors at three moments with a phase difference of 10*N / m between the power frequency cycles, the real-time frequency of the power system is obtained by using the phasor frequency measurement method; the real-time frequency of the power system is:

[0045]

[0046] Wherein, p=10*N / m, and 10*N / m is an integer, T s is the sampling interval, U re (i) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i, U im (i) is the imaginary part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i (i.e., the time corresponding to the i-th sampling number), U re (ip) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time ip, U im (ip) is the imaginary part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time ip, U re (i-2p) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i-2p, U im (i-2p) is the imaginary part of the phasor obtained by sinusoidal Fourier calculation of the low-frequency signal of the power system at time i-2p.

[0047] 4) Determine whether low-frequency oscillation occurs in the system.

[0048] The obtained real-time frequency value is averaged at N points to determine whether the average value is within the low-frequency range. When the average value is within the low-frequency range for a period of time exceeding a certain range, it is considered that the system has experienced low-frequency oscillation. Specifically, the method for calculating the N-point average value is: the real-time frequency corresponding to each sampling point in a power frequency cycle is added and divided by the number of sampling points to obtain the real-time frequency average value.

[0049] As an embodiment, the low frequency range is 0.198 Hz to 2.502 Hz. When the duration of the average value in the low frequency range exceeds 10 ms, it is considered that the system has low frequency oscillation.

[0050] To verify the effect of the low-frequency recognition method proposed in the present invention, the signal model is as follows:

[0051]

[0052] Among them, A0 is the effective value of signal 0, which is 100V, A1 is the effective value of signal 1, which is 0.05V, θ0 is the initial phase angle of signal 0, which is 20°, θ1 is the initial phase angle of signal 1, which is 50°, f0 is the frequency value of signal 0, which is 50Hz, and f1 is the frequency value of signal 1, which is the low-frequency oscillation amount superimposed on the power frequency. f1 is respectively 0.2Hz, 0.5Hz, 0.75Hz, 1.05Hz, 1.25Hz, 1.5Hz, 1.75Hz, 2.05Hz, and 2.5Hz. The low-frequency oscillation identification method proposed in the present invention is used to sample and identify the signal model under different values ​​of f1, and the following is obtained: Figure 2-Figure 10 The frequency value calculation results show that when 0.05% of the low-frequency oscillation is superimposed on the normal power frequency, the low-frequency oscillation can still be reliably identified by using the method of the present invention.

[0053] Compared with the methods used in the prior art, the calculation method is simpler, more reliable, easier to promote, and has greater practical application value.

[0054] Implementation Method of Low Frequency Oscillation Identification System for Power System

[0055] The present invention proposes a low-frequency oscillation identification system for an electric power system, comprising a processor. The processor adopts a low-frequency oscillation identification method to perform low-frequency oscillation identification.

[0056] The specific implementation process has been described in detail in the low-frequency oscillation identification method implementation method, and will not be repeated here.

Claims

1. A method for identifying low-frequency oscillations, characterized in that the steps include: 1) The original signal of the power system is sampled at a sampling rate of N sampling points per power frequency cycle, and the sampled signal is low-pass filtered to obtain a low-frequency signal; 2) Perform a sinusoidal Fourier transform calculation on the low-frequency signal at a fixed frequency of 1 Hz to obtain the phase quantities at three times with a phase difference of a*N / m power frequency cycles, where 20 ≥ a ≥ 5, m ≥ 3, and m is divisible by N; 3) Calculate the real-time frequency of the power system using the phasor frequency measurement method based on the real and imaginary parts of the phasors at three moments; 4) Calculate the real-time frequency average value. When the duration of the real-time frequency average value in the low-frequency range exceeds the time rating, it is considered that low-frequency oscillation occurs.

2. The low-frequency oscillation identification method according to claim 1, wherein: Performing a sinusoidal Fourier calculation on the low-frequency signal with a fixed frequency of 1 Hz, the expressions of the real and imaginary parts of the phasor are: Among them, i is the sampling number, u(i) is the low-frequency signal, U re is the real part of the phasor, U im is the imaginary part of the phasor, N1=50N.

3. The low frequency oscillation identification method according to claim 1, characterized in that: The expression of the real-time frequency of the power system is: Where p = a*N / m, T s is the sampling interval, U re (i) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i, U im (i) is the imaginary part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i, U re (ip) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time ip, U im (ip) is the imaginary part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time ip, U re (i-2p) is the real part of the phasor obtained by the sinusoidal Fourier calculation of the low-frequency signal of the power system at time i-2p, U im (i-2p) is the imaginary part of the phasor obtained by sinusoidal Fourier calculation of the low-frequency signal of the power system at time i-2p.

4. The low frequency oscillation identification method according to claim 1, wherein: The real-time frequency average value is calculated by adding the real-time frequencies corresponding to each sampling point within a power frequency cycle and dividing the sum by the number of sampling points to obtain the real-time frequency average value.

5. The low frequency oscillation identification method according to claim 1, characterized in that: The low-pass filtering method is to filter the sampled power system sampling signal through a Chebyshev filter.

6. The low frequency oscillation identification method according to claim 1, characterized in that: The low frequency range is 0.198 Hz to 2.502 Hz.

7. The low frequency oscillation identification method according to claim 1, characterized in that: The time is nominally 10 ms.

8. A low-frequency oscillation identification system for an electric power system, comprising a processor, characterized in that: The processor uses the low-frequency oscillation identification method according to any one of claims 1 to 7 to perform low-frequency oscillation identification.

Images