Wideband oscillation early warning method for power system
By using a broadband measurement platform and gradient enhancement tree model in the power system, real-time sampling and early warning, the accuracy and real-time problems of broadband oscillation monitoring of the power system are solved, and the accurate identification and early warning of broadband oscillation is achieved, and the power system is realized. Stable operation provides a basis.
Patent Information
- Application Number
- CN202410946141.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-05
AI Technical Summary
The prior art cannot effectively monitor the broadband oscillation of the power system accurately, real-time and quantitatively, affecting the stable operation of the power system.
By sampling the power grid in real time based on the broadband measurement platform, a relative power angle time data sequence is constructed, the damping ratio of the power system is calculated, and a broadband oscillation warning is performed based on the damping ratio, and a broadband oscillation prediction is carried out in combination with the gradient enhancement tree model to predict broadband oscillation in the future period.
It realizes accurate and real-time monitoring and early warning of wide-frequency oscillations, can identify the oscillation level, and provides a preventive basis for the stable operation of the power system.
Smart Images

Figure CN118899869A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to real-time monitoring of power grids, and in particular to a method for early warning of broadband oscillation of a power system. Background Art
[0002] With the proposal of "carbon peak" and "carbon neutrality", the proportion of new energy power generation systems in the power system continues to increase, and the penetration rate of power electronic equipment in the power system is also increasing. The power system is gradually showing "double highs" (that is, a high proportion of renewable energy and a high proportion of power electronic equipment) characteristics.
[0003] Unlike traditional power systems, in "double-high" power systems, due to the interaction between power electronic equipment and various components in the power grid, broadband oscillation events occur frequently. This broadband oscillation has the characteristics of time-varying and nonlinearity, and the oscillation frequency range is very wide. Broadband oscillation usually involves multiple electrical equipment in multiple regions, seriously affecting the stable operation of the power system and deteriorating the power quality. Therefore, it is very necessary to effectively monitor the broadband oscillation caused by the grid connection of new energy. At the same time, effective monitoring of broadband oscillation is an important basis for preventing and suppressing broadband oscillation.
[0004] Commonly used methods for monitoring broadband oscillations in power systems include the FFT algorithm. The resolution of the FFT algorithm is proportional to the length of the time window. To obtain more accurate monitoring results, a longer time window length is required, which is not conducive to real-time monitoring of broadband oscillations and cannot effectively determine the level of broadband oscillations. Summary of the invention
[0005] 1. Technical issues to be resolved
[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides a method for early warning of broadband oscillation in a power system, which can effectively overcome the defect of the prior art that the broadband oscillation cannot be accurately, real-timely and quantitatively monitored.
[0007] (II) Technical solution
[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:
[0009] A method for early warning of broadband oscillation of a power system, which performs real-time sampling of a power grid based on a broadband measurement platform to obtain a broadband measurement signal, and determines whether broadband oscillation occurs in the power system according to the broadband measurement signal;
[0010] In the case of broadband oscillation in the power system, a relative power angle time data series is constructed, and the damping ratio of the power system is calculated, and broadband oscillation warning is carried out according to the damping ratio of the power system.
[0011] Preferably, the real-time sampling of the power grid based on the broadband measurement platform to obtain a broadband measurement signal, and judging whether broadband oscillation occurs in the power system according to the broadband measurement signal, includes:
[0012] S11. Perform real-time sampling of the power grid based on a broadband measurement platform to obtain a waveform sequence of electrical signals in each period;
[0013] S12, performing parameter identification on the electrical signal waveform sequence in each time period to obtain the amplitude of harmonics and interharmonics in each time period;
[0014] S13. Compare the amplitudes of harmonics and interharmonics in each time period, and when the amplitude changes meet the preset conditions, determine that broadband oscillation occurs in the power system.
[0015] Preferably, in S12, parameter identification is performed on the electrical signal waveform sequence in each time period to obtain the amplitude of the harmonics and interharmonics in each time period, including:
[0016] S121, estimating the noise intensity of the electrical signal waveform sequence, and setting the window length according to the noise intensity;
[0017] S122, using the data window to intercept the electrical signal waveform sequence, and constructing a Hankel matrix, and solving the Hankel matrix to obtain a rough estimate of the electrical signal frequency;
[0018] S123, obtaining precise estimation values of electrical signal parameters based on the rough estimation value of the electrical signal frequency and the truncated electrical signal waveform sequence, that is, obtaining the amplitudes of harmonics and interharmonics in the corresponding time period of the electrical signal waveform sequence.
[0019] Preferably, the step of constructing a relative power angle time data sequence, calculating a damping ratio of the power system, and performing a broadband oscillation early warning according to the damping ratio of the power system comprises:
[0020] S21, constructing a relative power angle time data sequence, performing power spectrum analysis on the relative power angle time data sequence, and obtaining power spectrum energy distribution;
[0021] S22, selecting the frequency with the largest energy from the power spectrum energy distribution, and determining the corresponding time window length;
[0022] S23, constructing a second-order moment sample matrix according to the time window length, and calculating the numerical rank of the second-order moment sample matrix;
[0023] S24, performing a Prony operation according to the numerical rank of the second-order moment sample matrix to obtain the amplitude, phase, frequency and attenuation factor of the oscillation;
[0024] S25, calculating the damping ratio of the power system according to the amplitude, phase, frequency and attenuation factor of the oscillation;
[0025] S26. Provide a broadband oscillation warning based on the damping ratio of the power system.
[0026] Preferably, in S22, selecting a frequency with maximum energy from the power spectrum energy distribution and determining a corresponding time window length includes:
[0027] When the maximum power spectrum energy is concentrated in the frequency range of 0.1 to 2.5 Hz, the power system is judged to be in low-frequency oscillation;
[0028] When the maximum power spectrum energy is concentrated in the frequency range of 5-20 Hz, the power system is judged to be in subsynchronous oscillation;
[0029] When the maximum power spectrum energy is concentrated in the frequency range of 50-60 Hz, the power system is judged to be in synchronous oscillation;
[0030] The length of the corresponding time window is determined according to the oscillation type of the power system.
[0031] Preferably, in S23, a second-order moment sample matrix is constructed according to the time window length, and the numerical rank of the second-order moment sample matrix is calculated, including:
[0032] The singular value decomposition (SVD) is used to calculate the numerical rank of the second-order moment sample matrix, including:
[0033] The average value of the singular value ratio is calculated using the following formula
[0034]
[0035] Among them, R i is the ratio of the i-th singular value, S i is the i-th singular value, n is the number of singular values;
[0036] Count all the singular value ratios that are greater than the average value R of the singular value ratios, and use the largest subscript among these singular value ratios as the numerical rank of the second-order moment sample matrix.
[0037] Preferably, in S24, a Prony operation is performed according to the numerical rank of the second-order moment sample matrix to obtain the amplitude, phase, frequency and attenuation factor of the oscillation, including:
[0038] The mathematical model x(t) of the Prony operation is:
[0039]
[0040] Among them, A j is the amplitude of the jth complex exponential function, θ j is the phase of the jth complex exponential function, f jis the frequency of the jth complex exponential function, ε j is the attenuation factor of the jth complex exponential function, N is the number of complex exponential functions, and t is time.
[0041] Preferably, calculating the damping ratio of the power system according to the amplitude, phase, frequency and attenuation factor of the oscillation in S25 includes:
[0042] According to the amplitude, phase, frequency and attenuation factor of the oscillation, the damping ratio ζ of the power system is calculated using the following formula:
[0043]
[0044] Among them, ε is the attenuation factor of the oscillation, ω is the angular frequency of the oscillation, ω=2πf, and f is the frequency of the oscillation.
[0045] Preferably, performing a broadband oscillation warning according to the damping ratio of the power system in S26 includes:
[0046] When the damping ratio of the power system is 0.03<ζ<0.05, the power system is judged to be in a low-level broadband oscillation, a third-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent;
[0047] When the damping ratio of the power system is 0<ζ≤0.03, the power system is judged to be in a medium-level broadband oscillation, a second-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent;
[0048] When the damping ratio ζ of the power system is less than 0, the power system is judged to be in a high-level broadband oscillation, a first-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent.
[0049] Preferably, the method further includes using the collected historical broadband measurement signals to train the gradient boosting tree model, so that the gradient boosting tree model learns to predict the damping ratio of the power system in the future time period according to the broadband measurement signals in the past time period, and inputs the broadband measurement signals collected in real time into the trained gradient boosting tree model to predict the broadband oscillation of the power system in the future time period;
[0050] Among them, the gradient boosting tree model F(x) is expressed as follows:
[0051]
[0052] In the above formula, F 0 (x) is the initial gradient boosting tree model, F 0 (x) is a constant, T s (x) is the weak learner trained in the sth step, σ sis the learning rate of the s-th step training, M is the number of training iterations, and x is the independent variable.
[0053] (III) Beneficial effects
[0054] Compared with the prior art, the method for early warning of broadband oscillation in a power system provided by the present invention has the following beneficial effects:
[0055] 1) Based on the broadband measurement platform, the power grid is sampled in real time to obtain the electrical signal waveform sequence of each time period, and the parameters of the electrical signal waveform sequence of each time period are identified to obtain the amplitude of the harmonics and interharmonics in each time period. The amplitude of the harmonics and interharmonics in each time period is compared. When the amplitude change meets the preset conditions, it is judged that the power system has broadband oscillation, so that the amplitude of the harmonics and interharmonics in each time period can be obtained from the electrical signal waveform sequence of each time period, and the broadband oscillation can be accurately and real-time monitored according to the change of the amplitude of the harmonics and interharmonics in each time period;
[0056] 2) Construct a relative power angle time data sequence, perform power spectrum analysis on the relative power angle time data sequence, obtain power spectrum energy distribution, select the frequency with the largest energy from the power spectrum energy distribution, and determine the corresponding time window length, construct a second-order moment sample matrix according to the time window length, and perform Prony operation, calculate the damping ratio of the power system according to the Prony operation result, so as to accurately identify the broadband oscillation level of the power system according to the damping ratio of the power system, perform corresponding broadband oscillation warning, and obtain parameters such as amplitude, phase, frequency and attenuation factor that are helpful to further suppress broadband oscillation;
[0057] 3) The collected historical broadband measurement signals are used to train the gradient boosting tree model, so that the gradient boosting tree model can learn to predict the damping ratio of the power system in the future period based on the broadband measurement signals in the past period, and input the broadband measurement signals collected in real time into the trained gradient boosting tree model, so as to accurately predict the broadband oscillation of the power system in the future period, provide a basis for the prevention of broadband oscillations, and ensure the stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0059] Figure 1 It is a schematic diagram of the process of the present invention;
[0060] Figure 2 A schematic diagram of a flow chart for determining whether a wide-band oscillation occurs in a power system in the present invention;
[0061] Figure 3 It is a schematic diagram of the process of performing broadband oscillation early warning according to the damping ratio of the power system in the present invention. DETAILED DESCRIPTION
[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0063] A method for early warning of broadband oscillation in a power system, such as Figure 1 As shown, based on the broadband measurement platform, the power grid is sampled in real time to obtain a broadband measurement signal, and it is determined whether a broadband oscillation occurs in the power system according to the broadband measurement signal;
[0064] In the case of broadband oscillation in the power system, a relative power angle time data series is constructed, and the damping ratio of the power system is calculated, and broadband oscillation warning is carried out according to the damping ratio of the power system.
[0065] ① Based on the broadband measurement platform, the power grid is sampled in real time to obtain broadband measurement signals, and the broadband measurement signals are used to determine whether broadband oscillation occurs in the power system, such as Figure 2 As shown, including:
[0066] S11. Perform real-time sampling of the power grid based on a broadband measurement platform to obtain a waveform sequence of electrical signals in each period;
[0067] S12, performing parameter identification on the electrical signal waveform sequence in each time period to obtain the amplitude of harmonics and interharmonics in each time period;
[0068] S13. Compare the amplitudes of harmonics and interharmonics in each time period, and when the amplitude changes meet the preset conditions, determine that broadband oscillation occurs in the power system.
[0069] Specifically, in S12, parameter identification is performed on the electrical signal waveform sequence in each time period to obtain the amplitude of the harmonics and interharmonics in each time period, including:
[0070] S121, estimating the noise intensity of the electrical signal waveform sequence, and setting the window length according to the noise intensity;
[0071] S122, using the data window to intercept the electrical signal waveform sequence, and constructing a Hankel matrix, and solving the Hankel matrix to obtain a rough estimate of the electrical signal frequency;
[0072] S123, obtaining precise estimation values of electrical signal parameters based on the rough estimation value of the electrical signal frequency and the truncated electrical signal waveform sequence, that is, obtaining the amplitudes of harmonics and interharmonics in the corresponding time period of the electrical signal waveform sequence.
[0073] The above technical scheme performs real-time sampling of the power grid based on a broadband measurement platform, obtains the electrical signal waveform sequence of each time period, performs parameter identification on the electrical signal waveform sequence of each time period, obtains the amplitude of harmonics and interharmonics in each time period, compares the amplitude of harmonics and interharmonics in each time period, and judges that broadband oscillation occurs in the power system when the amplitude change meets the preset conditions, so that the amplitude of harmonics and interharmonics in each time period can be obtained from the electrical signal waveform sequence of each time period, and accurate and real-time monitoring of broadband oscillation can be achieved according to the change of the amplitude of harmonics and interharmonics in each time period.
[0074] ② Construct the relative power angle time data series and calculate the damping ratio of the power system. According to the damping ratio of the power system, a broadband oscillation warning is carried out. Figure 3 As shown, including:
[0075] S21, constructing a relative power angle time data sequence, performing power spectrum analysis on the relative power angle time data sequence, and obtaining power spectrum energy distribution;
[0076] S22, selecting the frequency with the largest energy from the power spectrum energy distribution, and determining the corresponding time window length;
[0077] S23, constructing a second-order moment sample matrix according to the time window length, and calculating the numerical rank of the second-order moment sample matrix;
[0078] S24, performing a Prony operation according to the numerical rank of the second-order moment sample matrix to obtain the amplitude, phase, frequency and attenuation factor of the oscillation;
[0079] S25, calculating the damping ratio of the power system according to the amplitude, phase, frequency and attenuation factor of the oscillation;
[0080] S26. Provide a broadband oscillation warning based on the damping ratio of the power system.
[0081] 1) In S22, the frequency with the largest energy is selected from the power spectrum energy distribution, and the corresponding time window length is determined, including:
[0082] When the maximum power spectrum energy is concentrated in the frequency range of 0.1 to 2.5 Hz, the power system is judged to be in low-frequency oscillation;
[0083] When the maximum power spectrum energy is concentrated in the frequency range of 5-20 Hz, the power system is judged to be in subsynchronous oscillation;
[0084] When the maximum power spectrum energy is concentrated in the frequency range of 50-60 Hz, the power system is judged to be in synchronous oscillation;
[0085] The length of the corresponding time window is determined according to the oscillation type of the power system.
[0086] 2) In S23, a second-order moment sample matrix is constructed according to the time window length, and the numerical rank of the second-order moment sample matrix is calculated, including:
[0087] The singular value decomposition (SVD) is used to calculate the numerical rank of the second-order moment sample matrix, including:
[0088] The average value of the singular value ratio is calculated using the following formula
[0089]
[0090] Among them, R i is the ratio of the i-th singular value, S i is the i-th singular value, n is the number of singular values;
[0091] Count the average values of all values greater than the singular value ratio The singular value ratios of , and the largest subscript among these singular value ratios is used as the numerical rank of the second-order moment sample matrix.
[0092] 3) In S24, a Prony operation is performed according to the numerical rank of the second-order moment sample matrix to obtain the amplitude, phase, frequency and attenuation factor of the oscillation, including:
[0093] The mathematical model x(t) of the Prony operation is:
[0094]
[0095] Among them, A j is the amplitude of the jth complex exponential function, θ j is the phase of the jth complex exponential function, f j is the frequency of the jth complex exponential function, ε j is the attenuation factor of the jth complex exponential function, N is the number of complex exponential functions, and t is time.
[0096] 4) In S25, the damping ratio of the power system is calculated based on the amplitude, phase, frequency and attenuation factor of the oscillation, including:
[0097] According to the amplitude, phase, frequency and attenuation factor of the oscillation, the damping ratio ζ of the power system is calculated using the following formula:
[0098]
[0099] Among them, ε is the attenuation factor of the oscillation, ω is the angular frequency of the oscillation, ω=2πf, and f is the frequency of the oscillation.
[0100] 5) S26 performs broadband oscillation warning according to the damping ratio of the power system, including:
[0101] When the damping ratio of the power system is 0.03<ζ<0.05, the power system is judged to be in a low-level broadband oscillation, a third-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent;
[0102] When the damping ratio of the power system is 0<ζ≤0.03, the power system is judged to be in a medium-level broadband oscillation, a second-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent;
[0103] When the damping ratio ζ of the power system is less than 0, the power system is judged to be in a high-level broadband oscillation, a first-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent.
[0104] The above technical scheme constructs a relative power angle time data sequence, performs power spectrum analysis on the relative power angle time data sequence, obtains the power spectrum energy distribution, selects the frequency with the largest energy from the power spectrum energy distribution, and determines the corresponding time window length, constructs a second-order moment sample matrix according to the time window length, and performs Prony operation, calculates the damping ratio of the power system according to the Prony operation result, so that the broadband oscillation level of the power system can be accurately identified according to the damping ratio of the power system, the corresponding broadband oscillation early warning can be carried out, and parameters such as amplitude, phase, frequency and attenuation factor that are helpful to further suppress broadband oscillation can be obtained.
[0105] ③ The technical solution of the present application also includes using the collected historical broadband measurement signals to train the gradient boosting tree model, so that the gradient boosting tree model learns to predict the damping ratio of the power system in the future period based on the broadband measurement signals in the past period, and inputs the broadband measurement signals collected in real time into the trained gradient boosting tree model to predict the broadband oscillation of the power system in the future period;
[0106] Among them, the gradient boosting tree model F(x) is expressed as follows:
[0107]
[0108] In the above formula, F 0 (x) is the initial gradient boosting tree model, F 0 (x) is a constant, Ts (x) is the weak learner trained in the sth step, σ s is the learning rate of the s-th step training, M is the number of training iterations, and x is the independent variable.
[0109] The above technical scheme uses the collected historical broadband measurement signals to train the gradient boosting tree model, so that the gradient boosting tree model learns to predict the damping ratio of the power system in the future time period based on the broadband measurement signals in the past time period, and inputs the broadband measurement signals collected in real time into the trained gradient boosting tree model, so as to accurately predict the broadband oscillation of the power system in the future time period, provide a basis for the prevention of broadband oscillations, and ensure the stable operation of the power system.
[0110] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for early warning of broadband oscillation in a power system, characterized by: Based on the broadband measurement platform, the power grid is sampled in real time to obtain broadband measurement signals, and the broadband measurement signals are used to determine whether broadband oscillation occurs in the power system. In the case of broadband oscillation in the power system, a relative power angle time data series is constructed, and the damping ratio of the power system is calculated, and broadband oscillation warning is carried out according to the damping ratio of the power system.
2. The method for early warning of broadband oscillation in a power system according to claim 1, characterized in that: The method of sampling the power grid in real time based on the broadband measurement platform to obtain a broadband measurement signal and judging whether broadband oscillation occurs in the power system according to the broadband measurement signal includes: S11. Perform real-time sampling of the power grid based on a broadband measurement platform to obtain a waveform sequence of electrical signals in each period; S12, performing parameter identification on the electrical signal waveform sequence in each time period to obtain the amplitude of harmonics and interharmonics in each time period; S13. Compare the amplitudes of harmonics and interharmonics in each time period, and when the amplitude changes meet the preset conditions, determine that broadband oscillation occurs in the power system.
3. The method for early warning of broadband oscillation in a power system according to claim 2, characterized in that: In S12, the electrical signal waveform sequence of each time period is parameter identified to obtain the amplitude of the harmonics and interharmonics in each time period, including: S121, estimating the noise intensity of the electrical signal waveform sequence, and setting the window length according to the noise intensity; S122, intercepting the electrical signal waveform sequence using the data window, and constructing a Hankel matrix, and obtaining a rough estimate of the electrical signal frequency according to the solution of the Hankel matrix; S123, obtaining precise estimation values of electrical signal parameters based on the rough estimation value of the electrical signal frequency and the truncated electrical signal waveform sequence, that is, obtaining the amplitudes of harmonics and interharmonics in the corresponding time period of the electrical signal waveform sequence.
4. The method for early warning of broadband oscillation in a power system according to claim 1, characterized in that: The relative power angle time data sequence is constructed, and the damping ratio of the power system is calculated, and a broadband oscillation early warning is performed according to the damping ratio of the power system, including: S21, constructing a relative power angle time data sequence, performing power spectrum analysis on the relative power angle time data sequence, and obtaining power spectrum energy distribution; S22, selecting the frequency with the largest energy from the power spectrum energy distribution, and determining the corresponding time window length; S23, constructing a second-order moment sample matrix according to the time window length, and calculating the numerical rank of the second-order moment sample matrix; S24, performing a Prony operation according to the numerical rank of the second-order moment sample matrix to obtain the amplitude, phase, frequency and attenuation factor of the oscillation; S25, calculating the damping ratio of the power system according to the amplitude, phase, frequency and attenuation factor of the oscillation; S26. Provide broadband oscillation warning based on the damping ratio of the power system.
5. The method for early warning of broadband oscillation in a power system according to claim 4, characterized in that: In S22, the frequency with the largest energy is selected from the power spectrum energy distribution, and the corresponding time window length is determined, including: When the maximum power spectrum energy is concentrated in the frequency range of 0.1 to 2.5 Hz, the power system is judged to be in low-frequency oscillation; When the maximum power spectrum energy is concentrated in the frequency range of 5-20 Hz, the power system is judged to be in subsynchronous oscillation; When the maximum power spectrum energy is concentrated in the frequency range of 50-60 Hz, the power system is judged to be in synchronous oscillation; The length of the corresponding time window is determined according to the oscillation type of the power system.
6. The method for early warning of broadband oscillation in a power system according to claim 4, characterized in that: In S23, a second-order moment sample matrix is constructed according to the time window length, and the numerical rank of the second-order moment sample matrix is calculated, including: The singular value decomposition (SVD) is used to calculate the numerical rank of the second-order moment sample matrix, including: The average value R of the singular value ratio is calculated using the following formula: Among them, R i is the ratio of the i-th singular value, S i is the i-th singular value, n is the number of singular values; Count the average values of all values greater than the singular value ratio The singular value ratios of , and the largest subscript among these singular value ratios is used as the numerical rank of the second-order moment sample matrix.
7. The method for early warning of broadband oscillation in a power system according to claim 4, characterized in that: In S24, the Prony operation is performed according to the numerical rank of the second-order moment sample matrix to obtain the amplitude, phase, frequency and attenuation factor of the oscillation, including: The mathematical model x(t) of the Prony operation is: Among them, A j is the amplitude of the jth complex exponential function, θ j is the phase of the jth complex exponential function, f j is the frequency of the jth complex exponential function, ε j is the attenuation factor of the jth complex exponential function, N is the number of complex exponential functions, and t is time.
8. The method for early warning of broadband oscillation in a power system according to claim 7, characterized in that: S25 calculates the damping ratio of the power system based on the amplitude, phase, frequency and attenuation factor of the oscillation, including: According to the amplitude, phase, frequency and attenuation factor of the oscillation, the damping ratio ζ of the power system is calculated using the following formula: Among them, ε is the attenuation factor of the oscillation, ω is the angular frequency of the oscillation, ω=2πf, and f is the frequency of the oscillation.
9. The method for early warning of broadband oscillation in a power system according to claim 4, characterized in that: S26 provides broadband oscillation warning based on the damping ratio of the power system, including: When the damping ratio of the power system is 0.03<ζ<0.05, the power system is judged to be in a low-level broadband oscillation, a third-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent; When the damping ratio of the power system is 0<ζ≤0.03, the power system is judged to be in a medium-level broadband oscillation, a second-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent; When the damping ratio ζ of the power system is less than 0, the power system is judged to be in a high-level broadband oscillation, a first-level warning is issued, and the corresponding oscillation amplitude, phase, frequency and attenuation factor are sent.
10. The method for early warning of broadband oscillation in a power system according to claim 1, characterized in that: The method also includes using the collected historical broadband measurement signals to train a gradient boosting tree model, so that the gradient boosting tree model learns to predict the damping ratio of the power system in the future period based on the broadband measurement signals in the past period, and inputting the broadband measurement signals collected in real time into the trained gradient boosting tree model to predict the broadband oscillation of the power system in the future period; Among them, the gradient boosting tree model F(x) is expressed as follows: In the above formula, F0(x) is the initial gradient boosting tree model, F0(x) is a constant, T s (x) is the weak learner trained in the sth step, σ s is the learning rate of the s-th step training, M is the number of training iterations, and x is the independent variable.
Citation Information
Cited By
Power system broadband measurement data processing method, device, equipment and medium
CN119646490A
Power system broadband measurement data processing method, device, equipment and medium
CN119646490B
