A Parameter Identification Method for Sub-Supersynchronous Oscillations Based on Elliptical Trajectory Fitting
By using an elliptical trajectory fitting method, the sampling values of the oscillation signal of the power system are directly used for time-domain calculation, which solves the problem of speed and accuracy in identifying sub-supersynchronous oscillation parameters in the power system and realizes real-time monitoring with low complexity.
Patent Information
- Application Number
- CN202611209968.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-11
- Publication Date
- 2026-11-10
AI Technical Summary
Existing technologies struggle to quickly and accurately identify subsynchronous oscillation parameters in power systems, especially with wideband oscillation signals. Furthermore, existing methods are computationally complex or rely on prior knowledge, making them unsuitable for real-time monitoring.
An elliptical trajectory fitting method is adopted. By acquiring the fundamental frequency of the oscillation signal and the instantaneous sampled values of the three-phase voltage, Clark transform and rotation factor processing are performed to fit the elliptical trajectory, calculate the sub-supersynchronous oscillation parameters, and directly use the original sampled values for time-domain calculation, avoiding the influence of frequency-domain transformation and prior parameters.
It enables rapid and accurate identification of sub-supersynchronous oscillation parameters, has high computational efficiency, is suitable for high sampling frequencies and short data windows, meets the requirements of speed and low latency for online real-time monitoring, and avoids problems of spectrum leakage and high computational complexity.
Smart Images

Figure CN123073712A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power system oscillation analysis technology, and more specifically, to a method for identifying subsynchronous oscillation parameters based on fitting synchronous phasor elliptical trajectories. Background Technology
[0002] With the rapid development of new energy power generation, high-voltage direct current transmission, and power electronic loads, the problem of sub / super synchronous oscillation (SSO) between power electronic devices and between them and the power grid in new power systems is becoming increasingly prominent. SSO-induced harmonics exhibit multimodal, rapidly time-varying, and wide-area propagation characteristics, making modal parameters difficult to identify. Furthermore, harmonic pollution propagates rapidly into the system; once SSO occurs, the oscillation waveforms often overlap, making it difficult to directly identify the location of the oscillation source. Therefore, real-time monitoring of oscillation parameters is crucial for rapidly identifying oscillation sources and ensuring power system stability.
[0003] In actual power grid operation, the oscillation modes of subsynchronous / supersynchronous oscillations (SSOs) are complex and diverse. The widely deployed power frequency phasor measurement units (PMUs) have limited dynamic bandwidth, making it difficult to fully capture the time-frequency evolution characteristics of broadband (several Hz to hundreds of Hz) oscillation signals. Furthermore, field operators lack sufficient experience in the rapid online identification of SSOs, relying heavily on offline manual analysis of fault recording data after incidents. There is a lack of a real-time, reliable, and applicable method for online identification of subsynchronous / supersynchronous oscillation parameters and oscillation source location in broadband oscillation scenarios.
[0004] Some researchers have improved the Fourier transform to obtain the spectrum of the synchronizing phasor, thereby identifying the oscillation parameters. However, this method requires the acquisition of a complete periodic signal to avoid spectral leakage, and the identification time is relatively long, making it difficult to meet the requirements of rapid response. Others have used the Prony analysis method to construct an exponential function linear model to fit the dynamic load characteristic curve, but the order of this algorithm model is difficult to determine, and excessively high order models can lead to a sharp increase in computational complexity and a significant decrease in computational speed.
[0005] Some researchers have used matched synchronous wavelet transform-based methods for estimating the instantaneous frequency of oscillations to identify parameters. However, this method suffers from endpoint effects at signal boundaries, and the selection of wavelet basis functions depends on prior knowledge, limiting its applicability. Others have analyzed the intrinsic modal components and time-frequency energy distribution characteristics of fault transient zero-sequence currents using variational mode decomposition (VMD) adaptive signal processing methods, but these methods lack universal parameter selection criteria and still suffer from high computational complexity. Summary of the Invention
[0006] The purpose of this invention is to overcome the defects and shortcomings of the prior art and provide a sub-supersynchronous oscillation parameter identification method based on elliptical trajectory fitting. This method is not affected by factors such as signal period integrity and prior parameter selection, has high computational efficiency, and can accurately and quickly identify sub-supersynchronous oscillation faults.
[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0008] A method for identifying subsynchronous oscillation parameters based on elliptical trajectory fitting includes the following steps:
[0009] During the subsynchronous oscillation of the power system, the fundamental frequency value and the instantaneous sampled value of the three-phase voltage in the oscillation signal are acquired;
[0010] Based on the sub-supersynchronous oscillation frequency relationship and the instantaneous value model of the oscillation signal, Clark transformation is performed on the instantaneous sampled values of the three-phase voltage to obtain the voltage components in the two-phase stationary coordinate system;
[0011] Based on the voltage components in the two-phase stationary coordinate system, an analytical signal is constructed. The analytical signal is multiplied by a rotation factor to suppress the fundamental component, and the processed instantaneous signal is obtained.
[0012] Multiple continuous sampling points are selected from the processed instantaneous signal, and an elliptical trajectory is fitted. Based on the analytical relationship between the elliptical geometric parameters and the oscillation parameters to be determined, the amplitude and initial phase of the fundamental component are calculated.
[0013] By subtracting the fundamental component from the analytical signal and fitting the elliptical trajectory again, the frequency, amplitude, and initial phase of the subsynchronous and supersynchronous components are calculated, thus realizing the identification of subsynchronous and supersynchronous oscillation parameters.
[0014] Furthermore, during the subsynchronous oscillation of a power system, the instantaneous values of voltage or current are considered as a superposition of multiple components, expressed as follows:
[0015] (1);
[0016] In the formula, This represents the amplitude of the fundamental component; The amplitude of the subsynchronous component; The amplitude of the supersynchronous component; The frequency of the fundamental component; The frequency of the subsynchronous component; The frequency of the supersynchronous component; This represents the initial phase of the fundamental component; This is the initial phase of the subsynchronous component; This represents the initial phase of the supersynchronous component;
[0017] Because subsynchronous oscillation and supersynchronous oscillation are coupled, the sum of their oscillation frequencies is equal to twice the fundamental frequency. The subsynchronous and supersynchronous oscillation frequencies have the following relationship:
[0018] (2);
[0019] Let w0 be the angular frequency of the fundamental component, w sub w is the angular frequency of the subsynchronous component. sup The angular frequency of the supersynchronous component is given by the angular frequency difference value. The angular frequencies of the three components have the following relationship: .
[0020] Furthermore, assume that f0 is known, i.e. Given that equation (1) can be rewritten, the instantaneous value model is:
[0021] (3);
[0022] In the formula, This represents the instantaneous value of the fundamental frequency component. This refers to the instantaneous value of the subsynchronous oscillation portion; This represents the instantaneous value of the supersynchronous oscillation component;
[0023] Multiply both sides of equation (3) by the rotation factor. The processed instantaneous value signal is obtained. :
[0024] (4);
[0025] In the formula, for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The value obtained.
[0026] Furthermore, the right side of the equation for the processed instantaneous signal consists of three parts: a fixed point, and a frequency of angular frequency. The forward-rotating subsynchronous component and the reverse-rotating supersynchronous component; the sum of the oscillating components forms an elliptical trajectory.
[0027] Taking the moment when the two oscillating components coincide as the reference point 0 rad, then since the positive and negative angular frequencies are respectively and − Therefore, within time dt, the subsynchronous part rotates by an angle θ, and θ = dt ⋅ The supersynchronous component has rotated by an angle −θ; therefore, the oscillation component is... ;
[0028] Oscillating components For an elliptical locus, the following equation holds:
[0029] (5);
[0030] In the formula, x is the real part of the oscillating component; y is the imaginary part of the oscillating component. denoted as , where b is the sum of the moduli of the sub-supersynchronous components; and denoted as , where b is the difference in the moduli of the sub-supersynchronous components.
[0031] Furthermore, the power system is a three-phase system, let the three-phase voltages be V. a V b V c Substituting the instantaneous value model into the three-phase system and performing Clark transformation, the voltage is projected onto the αβ axis to obtain the voltage on the αβ axis. , The analytic signal is constructed as follows:
[0032] (6);
[0033] Multiply both sides of equation (6) by the rotation factor. The instantaneous voltage signal after processing is obtained:
[0034] (7);
[0035] In the formula, for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The obtained values show that the trajectory characteristics are consistent between three-phase and single-phase, and the right side of the equation is the superposition of a point and an elliptical trajectory.
[0036] Furthermore, taking single-phase instantaneous values as an example, by processing the instantaneous values... After sampling five points consecutively, the coordinates of the center point of the ellipse (C1, C2) and the fundamental amplitude are determined. Initial phase with fundamental wave for:
[0037] (8);
[0038] The elliptical trajectory is considered as the superposition of two circular trajectories with opposite angular frequencies;
[0039] To make it easier to understand, the problem is simplified as follows:
[0040] Three consecutive interval sampling points on the ellipse are the corresponding two circular trajectories. The three corresponding consecutive interval sampling points are added together:
[0041] (9);
[0042] In the formula, X1, X2, and X3 are the upsampling points of the ellipse, and r1 and r2 are the radii of the circular trajectory, respectively. denoted as the initial phase of the circular trajectory, and w as the angular frequency of the circular trajectory.
[0043] Furthermore, assuming that the circular trajectory accumulates over time during sampling... Phase increment Expanding the trigonometric functions in equation (9), we get:
[0044] (10);
[0045] Equation (10) contains identical terms. To simplify the calculation, let the two identical terms be C and S. , Equation (10) simplifies to:
[0046] (11);
[0047] Continuing to expand the trigonometric functions of equation (11), we get:
[0048] (12);
[0049] remember , , , Equation (10) is finally simplified to:
[0050] (13).
[0051] Furthermore, by simultaneously solving the three equations in equation (13), we obtain:
[0052] (14);
[0053] Will , , , , , Substituting into equation (14), we get:
[0054] (15);
[0055] The angular frequency of the oscillation part in equation (4) Find:
[0056] (16).
[0057] Furthermore, the frequencies of the subsynchronous component and the supersynchronous component are:
[0058] (17);
[0059] In the formula, f s This refers to the sampling frequency of the power system.
[0060] Furthermore, the frequency f of the obtained subsynchronous component is... sub and the frequency f of the supersynchronous component sup ,as well as Substituting into equation (4), we obtain equation (18), and calculate the amplitude and phase of the subsynchronous component and the supersynchronous component:
[0061] (18);
[0062] Assume the selected sampling points are any segment of the entire time series being extracted, and that the selection does not begin from the first point of the series; let the initial point of the extracted time series be time zero; multiply each selected point by e. -jω0t deal with, The processing value for the first selected point. The processing value for the selected second point. This is the processing value for the selected (N-1)th point; The time from time zero to the first selected point; The time from time zero to the selected second point; The time from time zero to the selected (N-1)th point; only a minimum of three points are needed to solve for the frequency, amplitude, and phase of the subsynchronous and supersynchronous components in the instantaneous voltage or current values.
[0063] Compared with existing technologies, this invention proposes a rapid parameter identification method based on the intrinsic time-domain characteristics of the oscillation signal for subsynchronous oscillation fault waveforms in power systems. This method can achieve real-time extraction of oscillation frequency, amplitude, and attenuation coefficient without relying on frequency domain transformation. This invention directly uses the original sampled values for calculation, without involving Fourier transform or frequency domain integration processes, fundamentally avoiding the spectral leakage problem caused by improper window function selection in traditional methods, and is not constrained by the integrity of the signal period.
[0064] This invention, based on elliptical trajectory fitting theory, maps the instantaneous sampling points of a three-phase oscillation signal to a two-dimensional plane. Parameters are solved through the analytical relationship between elliptical parameters and oscillation frequency, ensuring complete utilization of the signal's time-domain information without the need for truncation or zero-padding. This invention requires only a few consecutive sampling points to complete a single parameter estimation, resulting in extremely low computational complexity. It is suitable for applications with high sampling frequencies and short data windows, meeting the engineering requirements for speed and low latency in online real-time monitoring of sub-supersynchronous oscillations. Attached Figure Description
[0065] Figure 1 This is a schematic diagram of a sub-supersynchronous oscillation parameter identification method based on elliptical trajectory fitting.
[0066] Figure 2 The results show the identification of the component parameters when the subsynchronous frequency varies in the range of [5, 45] Hz.
[0067] Figure 3 The results of parameter identification for each component when the subsynchronous amplitude varies in intervals of 1 within the range of [25, 100].
[0068] Figure 4 The results of parameter identification for each component when the initial phase angle of the subsynchronous phase changes in intervals of π / 12 within the range of [-π, π]. Detailed Implementation
[0069] The sub-supersynchronous oscillation parameter identification method based on elliptical trajectory fitting of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0070] Please see Figure 1 This invention discloses a method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting, comprising the following steps:
[0071] During the subsynchronous oscillation of the power system, the fundamental frequency value and the instantaneous sampled value of the three-phase voltage in the oscillation signal are acquired;
[0072] Based on the sub-supersynchronous oscillation frequency relationship and the instantaneous value model of the oscillation signal, Clark transformation is performed on the instantaneous sampled values of the three-phase voltage to obtain the voltage components in the two-phase stationary coordinate system;
[0073] Based on the voltage components in the two-phase stationary coordinate system, an analytical signal is constructed. The analytical signal is multiplied by a rotation factor to suppress the fundamental component, and the processed instantaneous signal is obtained.
[0074] Multiple continuous sampling points are selected from the processed instantaneous signal, and an elliptical trajectory is fitted. Based on the analytical relationship between the elliptical geometric parameters and the oscillation parameters to be determined, the amplitude and initial phase of the fundamental component are calculated.
[0075] By subtracting the fundamental component from the analytical signal and fitting the elliptical trajectory again, the frequency, amplitude, and initial phase of the subsynchronous and supersynchronous components are calculated, thus realizing the identification of subsynchronous and supersynchronous oscillation parameters.
[0076] Specifically, during the subsynchronous oscillation of a power system, the instantaneous values of voltage or current are considered as a superposition of multiple components, expressed as follows:
[0077] (1);
[0078] In the formula, This represents the amplitude of the fundamental component; The amplitude of the subsynchronous component; The amplitude of the supersynchronous component; The frequency of the fundamental component; The frequency of the subsynchronous component; The frequency of the supersynchronous component; This represents the initial phase of the fundamental component; This is the initial phase of the subsynchronous component; This represents the initial phase of the supersynchronous component.
[0079] Because subsynchronous oscillation and supersynchronous oscillation are coupled, the sum of their oscillation frequencies is equal to twice the fundamental frequency. The subsynchronous and supersynchronous oscillation frequencies have the following relationship:
[0080] (2);
[0081] Let w0 be the angular frequency of the fundamental component, w sub w is the angular frequency of the subsynchronous component. sup The angular frequency of the supersynchronous component is given by the angular frequency difference value. The angular frequencies of the three components have the following relationship: .
[0082] Assume f0 is known, that is Given that equation (1) can be rewritten as:
[0083] (3);
[0084] In the formula, This represents the instantaneous value of the fundamental frequency component. This refers to the instantaneous value of the subsynchronous oscillation portion; This represents the instantaneous value of the supersynchronous oscillation component;
[0085] Multiply both sides of equation (3) by the rotation factor. The processed instantaneous value signal is obtained. :
[0086] (4);
[0087] In the formula, for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The value obtained.
[0088] At this point, the right side of equation (4) consists of three parts: a fixed point, and angular frequency. The system consists of a forward-rotating subsynchronous component and a reverse-rotating supersynchronous component. The sum of these oscillating components forms an elliptical trajectory. Taking the moment when the two oscillating components coincide as a reference point of 0 rad, the positive and negative angular frequencies are respectively... and − Therefore, within time dt, the subsynchronous part rotates by an angle θ, and θ = dt ⋅ The supersynchronous component has rotated by an angle −θ; therefore, the oscillation component is... It is easy to prove. For an elliptical locus, the following equation holds:
[0089] (5);
[0090] In the formula, x is the real part of the oscillating component; y is the imaginary part of the oscillating component. denoted as , where b is the sum of the moduli of the sub-supersynchronous components; and denoted as , where b is the difference in the moduli of the sub-supersynchronous components.
[0091] The power system is a three-phase system. Let the three-phase voltages be V. a V b V c Substituting the instantaneous value model into the three-phase system and performing Clark transformation, the voltage is projected onto the αβ axis to obtain the voltage on the αβ axis. , The analytic signal is constructed as follows:
[0092] (6);
[0093] Similarly, both sides of equation (6) are multiplied by the rotation factor. The instantaneous voltage signal after processing is obtained:
[0094] (7);
[0095] In the formula, for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The obtained values show that the trajectory characteristics are consistent between three-phase and single-phase scenarios. The right side of the equation is a superposition of a point and an elliptical trajectory, so the subsequent processing steps can be considered the same.
[0096] Taking single-phase instantaneous value as an example, by processing the instantaneous value... By sampling five points consecutively, the coordinates of the ellipse's center (C1, C2) can be determined, making it easy to ascertain the fundamental wave amplitude. Initial phase with fundamental wave for:
[0097] (8);
[0098] The elliptical trajectory can be viewed as the superposition of two circular trajectories with opposite angular frequencies. For ease of understanding, the problem is simplified here: three consecutively spaced sampling points on the ellipse are the sums of the three corresponding consecutively spaced sampling points of the two circular trajectories.
[0099] (9);
[0100] In the formula, X1, X2, and X3 are the upsampling points of the ellipse; r1 and r2 are the radii of the circular trajectory, respectively. denoted as the initial phase of the circular trajectory; w is the angular frequency of the circular trajectory.
[0101] Assuming the circular trajectory accumulates over time during sampling Phase increment Expanding the trigonometric functions in equation (9) yields:
[0102] (10);
[0103] Equation (10) contains identical terms. To simplify the calculation, let the two identical terms be C and S. , Equation (10) simplifies to:
[0104] (11);
[0105] Continuing to expand the trigonometric functions of equation (11), we get:
[0106] (12);
[0107] remember , , , Equation (10) is finally simplified to:
[0108] (13);
[0109] Solving the three equations in equation (13) simultaneously, we get:
[0110] (14);
[0111] Will , , , , , Substituting into equation (14), we get:
[0112] (15);
[0113] At this point, the angular frequency of the oscillation part in equation (4) Find:
[0114] (16);
[0115] At this time, the frequencies of the subsynchronous component and the supersynchronous component are:
[0116] (17);
[0117] In the formula, f s This refers to the sampling frequency of the power system.
[0118] The frequency f of the subsynchronous component is obtained. sub and the frequency f of the supersynchronous component sup ,as well as Substituting into equation (4), we obtain equation (18), from which the amplitude and phase of the subsynchronous component and the supersynchronous component can be calculated:
[0119] (18);
[0120] Assume the selected sampling points are any segment of the entire time series being extracted, and that the selection does not begin from the first point of the series. Let the initial point of the extracted time series be time zero. Multiply each selected point by e. -jω0t deal with, The processing value for the first selected point. The processing value for the selected second point. This is the processing value for the selected (N-1)th point; The time from time zero to the first selected point; The time from time zero to the selected second point; The time from time zero to the selected (N-1)th point. A minimum of three points are needed to solve the problem, but more points can be selected depending on the required accuracy. In practical applications, 3-5 points are sufficient to solve for the frequency, amplitude, and phase of the subsynchronous and supersynchronous components in the instantaneous voltage or current values.
[0121] To demonstrate the correctness and effectiveness of the method proposed in this invention, an oscillating instantaneous signal was generated using the MATLAB platform, with the following specific parameters: f s =10kHz; f0=50Hz; x0=60; = / 4;f sub =20Hz; f sup =80Hz; x sub =30; x sup =20; = / 6; = / 4. f sub x varies in 0.5 Hz intervals within the range of [5, 45] Hz. sub Varying in intervals of 1 within the range [25, 100] It varies in intervals of π / 12 within the range of [-π, π].
[0122] Figure 2 , Figure 3 and Figure 4 The theoretical and identified values of each parameter under the three operating conditions are presented as comparison curves. As can be seen from the figures, within the range of parameter variation, the identified value curve and the theoretical value curve show good consistency, and the maximum relative error of each parameter is controlled within an acceptable engineering range, fully verifying the accuracy and applicability of the proposed method under different signal characteristics.
[0123] Compared to current methods that use improved Fourier transform to obtain the oscillation parameter identification of synchronous phasor spectrum, which require the acquisition of complete periodic signals to avoid spectrum leakage and have long identification time, this invention directly uses the original sampled values for time-domain calculation, without the need for a complete periodic data window, and can complete parameter estimation in milliseconds, meeting the speed requirements of online real-time monitoring.
[0124] Compared to the current method of using Prony analysis to fit dynamic load characteristics, which has problems such as difficulty in determining the order of the algorithm model and the tendency for excessively high order to lead to a surge in computational complexity, the analytical calculation method based on elliptical trajectory fitting in this invention does not require a preset model order, has extremely low computational cost, and only requires 3 to 5 sampling points to complete a single parameter estimation, effectively avoiding the computational burden brought by high-order models.
[0125] Compared with current methods for estimating the instantaneous frequency of oscillations based on matched synchronous wavelet transform, which suffer from endpoint effects at signal boundaries and limitations in applicability due to the reliance on prior knowledge for wavelet basis function selection, this invention directly uses sampling points to fit elliptical geometric parameters. This eliminates the need to select basis functions, avoids endpoint effects, and has a wider range of applications.
[0126] Compared with the current VMD adaptive signal processing method for sub-supersynchronous oscillation analysis, which lacks universal parameter selection criteria and has high computational complexity, making it unsuitable for real-time engineering applications, this invention adopts a direct time-domain calculation method. It does not require preset decomposition layers and penalty factors, and the calculation is simple and efficient, making it more suitable for online real-time monitoring scenarios in practical engineering.
[0127] The above description is a detailed description of the preferred embodiments of the present invention. However, the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications made under the technical spirit disclosed in the present invention should fall within the patent scope covered by the present invention.
Claims
1. A method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting, characterized in that, Includes the following steps: During the subsynchronous oscillation of the power system, the fundamental frequency value and the instantaneous sampled value of the three-phase voltage in the oscillation signal are acquired; Based on the sub-supersynchronous oscillation frequency relationship and the instantaneous value model of the oscillation signal, Clark transformation is performed on the instantaneous sampled values of the three-phase voltage to obtain the voltage components in the two-phase stationary coordinate system; Based on the voltage components in the two-phase stationary coordinate system, an analytical signal is constructed. The analytical signal is multiplied by a rotation factor to suppress the fundamental component, and the processed instantaneous signal is obtained. Multiple continuous sampling points are selected from the processed instantaneous signal, and an elliptical trajectory is fitted. Based on the analytical relationship between the elliptical geometric parameters and the oscillation parameters to be determined, the amplitude and initial phase of the fundamental component are calculated. By subtracting the fundamental component from the analytical signal and fitting the elliptical trajectory again, the frequency, amplitude, and initial phase of the subsynchronous and supersynchronous components are calculated, thus realizing the identification of subsynchronous and supersynchronous oscillation parameters.
2. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 1, characterized in that, During subsynchronous oscillations in a power system, the instantaneous values of voltage or current are considered as a superposition of multiple components, expressed as follows: (1); In the formula, This represents the amplitude of the fundamental component; The amplitude of the subsynchronous component; The amplitude of the supersynchronous component; The frequency of the fundamental component; The frequency of the subsynchronous component; The frequency of the supersynchronous component; This represents the initial phase of the fundamental component; This is the initial phase of the subsynchronous component; This represents the initial phase of the supersynchronous component; Because subsynchronous oscillation and supersynchronous oscillation are coupled, the sum of their oscillation frequencies is equal to twice the fundamental frequency. The subsynchronous and supersynchronous oscillation frequencies have the following relationship: (2); Let w0 be the angular frequency of the fundamental component, w sub w is the angular frequency of the subsynchronous component. sup The angular frequency of the supersynchronous component is given by the angular frequency difference value. The angular frequencies of the three components have the following relationship: .
3. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 2, characterized in that, Assume f0 is known, that is Given that equation (1) can be rewritten, the instantaneous value model is: (3); In the formula, This represents the instantaneous value of the fundamental frequency component; This refers to the instantaneous value of the subsynchronous oscillation portion; This represents the instantaneous value of the supersynchronous oscillation portion; Multiply both sides of equation (3) by the rotation factor. The processed instantaneous value signal is obtained. : (4); In the formula, for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The value obtained.
4. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 3, characterized in that, The right side of the equation for the processed instantaneous signal consists of three parts: a fixed point, and a frequency of angular frequency. The forward-rotating subsynchronous component and the reverse-rotating supersynchronous component; the sum of the oscillating components forms an elliptical trajectory. Taking the moment when the two oscillating components coincide as the reference point 0 rad, then since the positive and negative angular frequencies are respectively and − Therefore, within time dt, the subsynchronous portion rotates by an angle θ, and θ = dt ⋅ The supersynchronous component has rotated by an angle −θ; therefore, the oscillation component is... ; Oscillating components For an elliptical locus, the following equation holds: (5); In the formula, x is the real part of the oscillating component; y is the imaginary part of the oscillating component. denoted as , where b is the sum of the moduli of the sub-supersynchronous components; and denoted as , where b is the difference in the moduli of the sub-supersynchronous components.
5. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 3, characterized in that, The power system is a three-phase system. Let the three-phase voltages be V. a V b V c Substituting the instantaneous value model into the three-phase system and performing Clark transformation, the voltage is projected onto the αβ axis to obtain the voltage on the αβ axis. , The analytic signal is constructed as follows: (6); Multiply both sides of equation (6) by the rotation factor. The instantaneous value signal after voltage processing is obtained: (7); In the formula, for Multiply by the rotation factor The value obtained; for Multiply by the rotation factor The obtained values show that the trajectory characteristics are consistent between three-phase and single-phase, and the right side of the equation is the superposition of a point and an elliptical trajectory.
6. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 5, characterized in that, Taking single-phase instantaneous value as an example, by processing the instantaneous value... After sampling five points consecutively, the coordinates of the center point of the ellipse (C1, C2) and the fundamental amplitude are determined. Initial phase with fundamental wave for: (8); The elliptical trajectory is considered as the superposition of two circular trajectories with opposite angular frequencies; To make it easier to understand, the problem is simplified as follows: Three consecutive interval sampling points on the ellipse are the corresponding two circular trajectories. The three corresponding consecutive interval sampling points are added together: (9); In the formula, X1, X2, and X3 are the upsampling points of the ellipse, and r1 and r2 are the radii of the circular trajectory, respectively. denoted as the initial phase of the circular trajectory, and w as the angular frequency of the circular trajectory.
7. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 6, characterized in that, Assuming the circular trajectory accumulates over time during sampling Phase increment Expanding the trigonometric functions in equation (9), we get: (10); Equation (10) contains identical terms. To simplify the calculation, let the two identical terms be C and S. , Equation (10) simplifies to: (11); Continuing to expand the trigonometric functions of equation (11), we get: (12); remember , , , Equation (10) is finally simplified to: (13)。 8. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 7, characterized in that, Solving the three equations in equation (13) simultaneously, we get: (14); Will , , , , , Substituting into equation (14), we get: (15); The angular frequency of the oscillation part in equation (4) Find: (16)。 9. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 8, characterized in that, The frequencies of the subsynchronous component and the supersynchronous component are: (17); In the formula, f s This refers to the sampling frequency of the power system.
10. The method for identifying sub-supersynchronous oscillation parameters based on elliptical trajectory fitting according to claim 9, characterized in that, The frequency f of the subsynchronous component is obtained. sub and the frequency f of the supersynchronous component sup ,as well as Substituting into equation (4), we obtain equation (18), and calculate the amplitude and phase of the subsynchronous component and the supersynchronous component: (18); Assume the selected sampling points are any segment of the entire time series being extracted, and that the selection does not begin from the first point of the series; let the initial point of the extracted time series be time zero; multiply each selected point by e. -jω0t deal with, The processing value for the first selected point. The processing value for the selected second point. This is the processing value for the selected (N-1)th point; The time from time zero to the first selected point; The time from time zero to the selected second point; The time from time zero to the selected (N-1)th point; With a minimum of only three points, the frequency, amplitude, and phase of the subsynchronous and supersynchronous components in the instantaneous values of voltage or current can be determined.
