Method and system for screening sub-synchronous oscillation curve, and storage medium and electronic device
By combining a custom time window length and a support vector machine classifier with a time-frequency domain wavelet packet decomposition method, subsynchronous oscillation curves in power systems can be quickly screened out. This solves the problems of high computational load and noise impact in existing technologies, achieving efficient and accurate oscillation curve screening and ensuring power grid safety.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- CHINA ELECTRIC POWER RESEARCH INSTITUTE CO LTD
- Filing Date
- 2025-12-05
- Publication Date
- 2026-06-04
AI Technical Summary
Existing technologies involve large computational loads and long analysis times when screening subsynchronous oscillation curves. Furthermore, the Prony algorithm is prone to failure in the presence of noise, leading to calculation errors or fitting failures, making it difficult to quickly and accurately screen subsynchronous oscillation phenomena in power systems.
A sequence of electromagnetic transient simulation phasor data points is obtained by using a custom time window length. A classifier is trained by a support vector machine to quickly filter oscillation behavior. Noise is filtered out by combining the time-frequency domain wavelet packet decomposition method. The frequency and amplitude of the interharmonic components are obtained. The subsynchronous oscillation curve is determined by using the sensitive frequency band and the energy accumulation threshold.
It enables rapid and accurate screening of dangerous subsynchronous oscillation data, improving screening efficiency and accuracy, and allowing for timely response to oscillation divergence, thus ensuring the safety and stability of the power grid.
Smart Images
Figure CN2025140461_04062026_PF_FP_ABST
Abstract
Description
Methods and systems for screening subsynchronous oscillation curves, storage media, and electronic devices
[0001] This disclosure claims priority to Chinese patent application No. 202411699037.8, filed on November 26, 2024, the entire contents of which are incorporated herein by reference. Technical Field
[0002] This disclosure relates to the field of power system safety and stability analysis, and more specifically, to a method and system for screening subsynchronous oscillation curves, a storage medium, and electronic equipment. Background Technology
[0003] With the rapid development of new energy technologies, large-capacity, centralized grid-connected wind / solar power bases, utilizing ultra-high voltage (UHV) or flexible DC transmission, are currently the dominant form of large-scale renewable energy consumption. In areas with large-scale renewable energy aggregation, grid connection requires the installation of numerous power electronic devices. The fast control devices used in these power electronic devices require current phase-locked loops, which can introduce subsynchronous / supersynchronous harmonic components into the power signal during power surges or drops, causing subsynchronous oscillations in the power system. Therefore, the selection of subsynchronous oscillation curves is crucial for the safe and stable operation of the power grid. Summary of the Invention
[0004] On one hand, embodiments of this disclosure provide a method for screening subsynchronous oscillation curves, including:
[0005] Based on the custom time window length, obtain the sequence of amplitude data points of electromagnetic transient simulation phasors to be determined for multiple consecutive time windows;
[0006] Input the sequence of amplitude data points to be determined into a pre-established classifier, and output the category of the sequence of amplitude data points to be determined. The category of the sequence of amplitude data points to be determined includes oscillating behavior and no oscillating behavior.
[0007] The time-frequency domain wavelet packet decomposition method is used to analyze the sequence of data points with oscillating behavior and the amplitude to be determined, so as to obtain the frequency and amplitude of the interharmonic components to be screened.
[0008] The amplitude percentage of the interharmonic component to be screened is compared with a predetermined amplitude threshold, and the interharmonic component to be screened after noise is filtered out is taken as the remaining interharmonic component. The amplitude percentage data refers to the percentage of the amplitude of the interharmonic component to the fundamental frequency.
[0009] The subsynchronous oscillation curve of the output is determined based on the frequency and amplitude percentage of the remaining interharmonic components.
[0010] On the other hand, embodiments of this disclosure provide a system for screening subsynchronous oscillation curves, the system comprising:
[0011] The data acquisition module is used to acquire a sequence of amplitude data points of electromagnetic transient simulation phasors to be determined for multiple consecutive time windows, based on a custom time window length.
[0012] The data classification module is used to input the sequence of amplitude data points to be determined into a pre-established classifier and output the category of the sequence of amplitude data points to be determined. The categories of the sequence of amplitude data points to be determined include those with oscillatory behavior and those without oscillatory behavior.
[0013] The data analysis module is used to employ time-frequency domain wavelet packet decomposition.
[0014] The method analyzes the sequence of amplitude data points with oscillating behavior to obtain the frequency and amplitude of the interharmonic components to be screened.
[0015] The noise filtering module is used to compare the amplitude percentage of the interharmonic component to be screened with a predetermined amplitude threshold, and to take the interharmonic component to be screened after noise filtering as the remaining interharmonic component. The amplitude percentage data refers to the percentage of the amplitude of the interharmonic component to the fundamental frequency.
[0016] The result output module is used to determine the output subsynchronous oscillation curve based on the frequency and amplitude percentage of the remaining interharmonic components.
[0017] In another aspect, this disclosure provides a computer-readable storage medium storing a computer program for performing the methods described in one aspect of this disclosure.
[0018] In another aspect, an electronic device is provided, the electronic device comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the method described in one aspect of this disclosure. Attached Figure Description
[0019] Exemplary embodiments of this disclosure can be more fully understood by referring to the following figures:
[0020] Figure 1 is a flowchart of a method for screening subsynchronous oscillation curves according to some embodiments.
[0021] Figure 2 is a schematic diagram of the structure of a system for screening subsynchronous oscillation curves according to some embodiments.
[0022] Figure 3 is a schematic diagram of the structure of an electronic device according to some embodiments. Detailed Implementation
[0023] Exemplary embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, the present disclosure may be embodied in many different forms and is not limited to the embodiments described herein. These embodiments are provided to fully and completely disclose the present disclosure and to fully convey its scope to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the drawings is not intended to limit the present disclosure. In the drawings, the same units / elements are referred to by the same reference numerals.
[0024] Unless otherwise stated, the terms used herein (including technical terms) have their common meaning as understood by one of ordinary skill in the art. Furthermore, it is understood that terms defined in commonly used dictionaries should be understood to have a meaning consistent with the context of their relevant field, and not to be interpreted as having an idealized or overly formal meaning.
[0025] With the rapid development of new energy technologies, large-capacity, centralized grid-connected wind / solar power bases, utilizing ultra-high voltage (UHV) or flexible DC transmission, are currently the dominant form of large-scale renewable energy consumption. In areas with large-scale renewable energy aggregation, grid connection requires the installation of numerous power electronic devices. The fast control devices used in these power electronic devices require current phase-locked loops, which can introduce subsynchronous / supersynchronous harmonic components into the power signal during power surges or drops, causing subsynchronous oscillations in the power system. Therefore, the selection of subsynchronous oscillation curves is crucial for the safe and stable operation of the power grid.
[0026] Some techniques primarily employ two methods to screen subsynchronous oscillation curves. One method involves performing frequency domain spectral analysis on a large number of data points, identifying subsynchronous oscillations based on the frequency, amplitude, and frequency of consecutive occurrences of the resulting interharmonic components. However, this approach requires continuous analysis for each oscillation curve, resulting in a large overall computational load and long analysis time, leaving room for improvement in the speed of subsynchronous oscillation screening. The other method uses Prony analysis. The Prony algorithm fits a mathematical model of equally spaced sampled data using a complex exponentially decaying linear combination. With appropriate extensions, it can directly estimate the frequency, attenuation factor, amplitude, and phase of a given signal, providing a more comprehensive representation of transient signal characteristics than traditional Fourier analysis. It shows promising application prospects in power system signal analysis, especially low-frequency oscillation analysis. The Prony algorithm can extract valuable information from uniformly sampled signals, decomposing a series of exponentially decaying sine curves. However, the Prony method assumes that the model is a combination of exponential functions with arbitrary amplitude, phase, frequency, and attenuation factor, i.e., composed of decaying sine / cosine signal components. This algorithm requires calculation of least squares fitting and finding complex roots of higher-order algebraic equations. When noise is embedded in the signal, the Prony solution process may fail due to ill-conditioned equations. If the input data is too long, or the estimated number of effective mode terms is too large, calculation errors or fitting failures may occur.
[0027] The following is a detailed description with reference to the embodiments.
[0028] Example 1: Exemplary Method
[0029] Figure 1 is a flowchart of a method for screening subsynchronous oscillation curves according to some embodiments. As shown in Figure 1, the method for screening subsynchronous oscillation curves according to embodiments of the present disclosure includes steps S101-S105.
[0030] In S101, based on the custom time window length, a sequence of amplitude data points to be determined for electromagnetic transient simulation phasors of multiple consecutive time windows is obtained.
[0031] This disclosure does not impose any special restrictions on electromagnetic transient simulation phasors; both current phasors and voltage phasors are acceptable.
[0032] In some embodiments, before acquiring the amplitude data point sequence of multiple consecutive time windows of electromagnetic transient simulation phasors according to a custom time window length, a classifier for classifying the amplitude data point sequence of electromagnetic transient simulation phasors to be determined is included. The classifier for classifying the amplitude data point sequence of electromagnetic transient simulation phasors to be determined includes:
[0033] Based on a custom time window length, obtain a sequence of historical amplitude data points for electromagnetic transient simulation phasors across multiple consecutive time windows;
[0034] According to the custom dimension number N, the discrete wavelet transform method is used to represent the historical amplitude data point sequence of the electromagnetic transient simulation phasor of each time window of multiple time windows as historical coordinate points in N-dimensional space. The number of decomposition layers of the discrete wavelet transform method is equal to the custom dimension number N.
[0035] The categories of historical coordinate points in the N-dimensional space are determined by using a preset coordinate point category criterion, and corresponding category labels are assigned according to the categories of historical coordinate points. The categories of historical coordinate points include those with oscillation behavior and those without oscillation behavior.
[0036] The classifier is generated by training the historical coordinate points and their corresponding category labels in the N-dimensional space using a support vector machine.
[0037] In this embodiment of the present disclosure, the time window length for collecting historical amplitude data points of electromagnetic transient simulation phasors during classifier training is the same as the time window length for collecting amplitude data points to be determined when the classifier is generated and used to classify unknown oscillation data. The custom dimension N is adjusted according to the complexity of the power grid, typically not less than 2, and is used as the decomposition level of the discrete wavelet transform method employed.
[0038] The main principle of using support vector machines (SVMs) to train a classifier is to select a kernel function, find the support vectors for each of the two classes' historical coordinate points in N-dimensional space, and determine the boundary between historical coordinate points exhibiting oscillatory behavior and those without, based on the criterion that maximizes the sum of the distances from the support vectors of both classes to the boundary. By establishing a classifier, it is possible to quickly filter the amplitude data point sequences of electromagnetic transient simulation phasors exhibiting oscillatory behavior, thus providing a better data source for further in-depth analysis of more dangerous oscillations.
[0039] In some embodiments, representing the sequence of historical amplitude data points of the electromagnetic transient simulation phasor for each time window as historical coordinate points in N-dimensional space using the discrete wavelet transform method according to a custom dimension N includes:
[0040] The discrete wavelet transform method is used to decompose and reconstruct the historical amplitude data point sequence of the electromagnetic transient simulation phasor in each window, and to determine the wavelet decomposition coefficients d of the nth level corresponding to the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window across multiple time windows. in , 1≤i≤I, where I is the total number of time windows, 1≤n≤N;
[0041] According to the wavelet decomposition coefficients d in Generate the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window across multiple time windows, corresponding to the historical coordinate points (d) in N-dimensional space. i1 ,d i2 ,…,d in ,…,d iN ).
[0042] In this embodiment of the disclosure, assuming that the custom dimension N is equal to 5, the discrete wavelet transform method is used to decompose and reconstruct the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window of multiple time windows, so that its corresponding historical coordinate point (d) in 5-dimensional space can be obtained. i1 ,d i2 ,d i3 ,d i4 ,d i5 ).
[0043] In some embodiments, the coordinate point category criterion is:
[0044] When the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window of multiple time windows corresponds to the historical coordinate point in the N-dimensional space (d... i1 ,d i2 ,…,d in ,…,d iNWhen ), if d ia With d ib When the absolute value of all differences is less than d0, the historical coordinate point is classified as having no oscillation behavior; otherwise, in d... ia With d ib If the absolute value of the difference is greater than or equal to d0, the category of the historical coordinate point is oscillating behavior, where d0 is a positive number less than 1, 1≤i≤I, 1≤a, b≤N, and a≠b.
[0045] According to research, when the curve of an electromagnetic transient simulation phasor in a time window does not oscillate significantly, the wavelet decomposition coefficients of each layer corresponding to the historical coordinate points in the N-dimensional space will be very close. Therefore, by determining that the wavelet decomposition coefficients of any two layers are less than a set constant, it is possible to relatively accurately determine whether the curve of an electromagnetic transient simulation phasor in a time window has oscillating behavior.
[0046] In some embodiments, after determining the category of historical coordinate points in the N-dimensional space using a preset coordinate point category criterion, the method includes determining an amplitude threshold for filtering noise in a window exhibiting oscillating behavior. Determining the amplitude threshold for filtering noise in a window exhibiting oscillating behavior includes:
[0047] The time-frequency domain wavelet packet decomposition method is used to analyze the sequence of historical amplitude data points corresponding to historical coordinate points with oscillating behavior, obtain the amplitude of the corresponding historical interharmonic components, and calculate the amplitude percentage of the historical interharmonic components.
[0048] According to the set interharmonic component classification rules, the category of the historical interharmonic component is determined based on the amplitude percentage of the historical interharmonic component, and a corresponding category label is added. The categories of the historical interharmonic components include those that need attention and those that do not need attention.
[0049] Based on the amplitude percentage and category label of the historical interharmonic components, a support vector machine learner is trained to obtain the boundary point on a one-dimensional number axis for the amplitude percentage data of the classes that need attention and the classes that do not need attention. The boundary point is the determined amplitude threshold.
[0050] It is important to note that when a new oscillation mode emerges, in order to make the amplitude threshold training more effective, the support vector machine learner should be updated in a timely manner when both the data points of the attention and non-attention classes exceed a certain value.
[0051] In some embodiments, the interharmonic component classification rules include:
[0052] Historical interharmonic components with an amplitude percentage below L are defined as the no-concern category;
[0053] Historical interharmonic components with an amplitude percentage higher than M% are defined as the class requiring attention, L. <M。
[0054] For historical interharmonic components that do not meet the above two definitions, if their amplitude percentage is higher than K times the average amplitude percentage of all historical interharmonic components in the same time window, the category is defined as a class that needs attention; otherwise, the category is defined as a class that does not need attention, where K is a natural number.
[0055] In some embodiments, the wavelet mother function used in the discrete wavelet transform method or the time-frequency domain wavelet packet decomposition method is a third-order Daubechies series wavelet, whose expression is: H(ω)=0.0249086-0.0604169e -iω -0.095467e -2iω +0.325186e -3iω
[0056] ω is the amplitude point sequence of the input electromagnetic transient simulation phasor with oscillating behavior, and H(ω) is the amplitude and frequency of the interharmonic components obtained after hierarchical decomposition of the input.
[0057] In some technologies, there are many wavelet mother functions that can be selected for discrete wavelet transform methods or time-frequency domain wavelet packet decomposition methods, such as Haar wavelets, Coiflet series wavelets, Symlets series wavelets, Morlets series wavelets, and Mexican Hat series wavelets. In this embodiment of the present disclosure, the third-order Daubechies series wavelet is selected and further refined into the expression in the above embodiment of the present disclosure. This is because the Daubechies series wavelet supports wavelet packet transform, has a low vanishing moment order, and a high oscillation frequency feature recognition rate compared with other wavelets. The comparison results are shown in Table 1.
[0058] As can be seen from the comparison in Table 1, the Daubechies series wavelets, as wavelet mother functions, have significant advantages over other wavelets when compared in three dimensions: wavelet packet transform, vanishing moment divisor, and oscillation frequency recognition ratio.
[0059] In S102, the sequence of amplitude data points to be determined is input into a pre-established classifier, which outputs the category of the sequence of amplitude data points to be determined. The category of the sequence of amplitude data points to be determined includes oscillating behavior and no oscillating behavior.
[0060] In S103, the time-frequency domain wavelet packet decomposition method is used to analyze the data point sequence of amplitude to be determined with oscillating behavior, and to obtain the frequency and amplitude of the interharmonic components to be screened.
[0061] In S104, the amplitude percentage of the interharmonic component to be screened is compared with a predetermined amplitude threshold, and the interharmonic component to be screened after noise filtering is taken as the remaining interharmonic component. The amplitude percentage refers to the percentage of the amplitude of the interharmonic component relative to the fundamental frequency.
[0062] In some embodiments, comparing the amplitude percentage of the interharmonic components to be screened with a predetermined amplitude threshold, and using the noise-filtered interharmonic components as the remaining interharmonic components, includes:
[0063] The amplitude percentage of the interharmonic components to be screened is compared with a predetermined amplitude threshold, and the interharmonic components to be screened whose amplitude percentage is less than the amplitude threshold are filtered out as noise.
[0064] The remaining part of the interharmonic components to be screened after filtering out noise is taken as the remaining interharmonic components.
[0065] In this embodiment of the disclosure, by filtering out interharmonic components with amplitude percentages and amplitude thresholds as noise, interharmonic components with small amplitude percentages that are not harmful due to noise can be excluded, while interharmonic components with large amplitudes and high risk can be given special attention, thereby improving the screening efficiency.
[0066] In S105, the subsynchronous oscillation curve of the output is determined based on the frequency and amplitude percentage of the remaining interharmonic components.
[0067] In some embodiments, determining the output subsynchronous oscillation curve based on the frequency and amplitude percentage of the remaining interharmonic components includes:
[0068] S1. Based on the frequency of each interharmonic component in the sensitive frequency band and the remaining interharmonic components, determine the energy accumulation benchmark threshold and energy accumulation weight value of each interharmonic component. The sensitive frequency band is a frequency range composed of the upper limit and lower limit values obtained by adding and subtracting a custom frequency step size from the difference between the power frequency and the motor modal frequency during electromagnetic transient simulation. Energy accumulation refers to the percentage of amplitude of the interharmonic component integrated over time. The first energy accumulation benchmark threshold of the interharmonic component with frequency in the sensitive frequency band is less than the second energy accumulation benchmark threshold of the interharmonic component with frequency not in the sensitive frequency band. The first energy accumulation weight value of the interharmonic component with frequency in the sensitive frequency band is greater than the second energy accumulation weight value of the interharmonic component with frequency not in the sensitive frequency band. The sum of the first energy accumulation weight value and the second energy accumulation weight value is between the frequency step size and twice the frequency step size. The first energy accumulation weight value is k times the second energy accumulation weight value, where k is a positive number not less than 3.
[0069] S2. According to the set time interval, calculate the cumulative energy value of each interharmonic component in the j-th time interval, and calculate the cumulative weighted sum of energy in the j-th time interval based on the cumulative energy weight value of each interharmonic component and the cumulative energy value in the j-th time interval. j is a natural number and the initial value of j is 1.
[0070] S3. When the cumulative energy value of each interharmonic component in the j-th time interval is less than its corresponding cumulative energy analysis threshold, and the cumulative weighted sum of energy in the j-th time interval is less than the set cumulative weighted sum of energy analysis threshold, execute S4; otherwise, execute S7. The cumulative weighted sum of energy analysis threshold is any positive integer between [5, 50]. When j = 1, the cumulative energy analysis threshold is equal to the cumulative energy reference threshold, and the cumulative energy value and the cumulative weighted sum of energy in the (j-1)-th time interval are equal to the cumulative energy value and the cumulative weighted sum of energy in the (j-1)-th time interval, respectively.
[0071] S4. Calculate the cumulative energy change rate of each interharmonic component based on the cumulative energy value of each interharmonic component in the j-th time interval and the (j-1)-th time interval, and calculate the weighted cumulative energy change rate based on the weighted sum of the cumulative energy values in the j-th time interval and the (j-1)-th time interval.
[0072] S5. Based on the set rate of change gradient threshold mapping relationship table, determine the first gradient threshold interval corresponding to the j-th cumulative energy change rate of each interharmonic component, and the second gradient threshold interval corresponding to the j-th cumulative weighted energy change rate. The first gradient threshold interval is generated by setting P1 gradient thresholds in ascending order of energy cumulative change rate, and making two adjacent gradient thresholds form a gradient threshold interval, resulting in P1-1 value intervals arranged in ascending order. The second gradient threshold interval is generated by setting P2 gradient thresholds in ascending order of energy cumulative weighted change rate, and making two adjacent gradient thresholds form a gradient threshold interval, resulting in P2-1 value intervals arranged in ascending order.
[0073] S6. When the first gradient threshold interval of the j-th cumulative energy change rate is greater than the first gradient threshold interval of the (j-1)-th cumulative energy change rate, or when the second gradient threshold interval corresponding to the j-th cumulative weighted energy change rate is greater than the second gradient threshold interval corresponding to the (j-1)-th cumulative energy change rate, according to the set gradient threshold energy cumulative threshold mapping relationship table, the energy cumulative threshold corresponding to the first gradient threshold interval of the (j-1)-th cumulative energy change rate is used as the energy cumulative analysis threshold for the (j+1)-th time interval. Otherwise, the energy cumulative threshold corresponding to the first gradient threshold interval of the j-th cumulative energy change rate is used as the updated energy cumulative analysis threshold. Let j = j + 1, and execute S2. When j = 1, the first gradient threshold interval of the (j-1)-th cumulative energy change rate is equal to the first gradient threshold interval of the j-th cumulative energy change rate.
[0074] S7. Output the interharmonic components from the first time interval to the j-th time interval as the selected sub-frequency oscillation curves.
[0075] In this embodiment, when the generator set modal frequency is 20Hz, the complementary characteristic frequency on the power system side is 30Hz. Considering the error, the frequency step size is set to 1.5Hz, thus the sensitive frequency range is 28.5Hz to 31.5Hz. Within the sensitive frequency range (28.5Hz to 31.5Hz), the first energy accumulation threshold for interharmonic components is correspondingly reduced to 30, while the second energy accumulation threshold for interharmonic components in the non-sensitive frequency range is higher than 30. An analysis threshold is set for the weighted sum of the energy accumulation of all coexisting interharmonic components. The first energy accumulation weight value for interharmonic components within the sensitive frequency range is set as a1, and the second energy accumulation weight value for interharmonic components in the non-sensitive frequency range is set as a2. According to the assignment method of this embodiment, a1 can be selected as 1.5, and a2 as 0.5, thus satisfying both 1.5 < a1 + a2 < 1.5 * 2 and a1 = 3a2. The cumulative weighted sum of energy refers to the sum of the product of the first cumulative weighted value of energy and the cumulative energy value of the interharmonic components within the sensitive frequency range and the product of the second cumulative weighted value of energy and the cumulative energy value of the interharmonic components within the non-sensitive frequency range.
[0076] In this embodiment, thresholds for the cumulative energy change rate and the weighted cumulative energy change rate are pre-set according to a gradient, forming a gradient threshold interval. This establishes a correspondence between the cumulative energy change rate and the weighted cumulative energy change rate and the gradient threshold interval, thereby creating a gradient threshold mapping table. During the screening process, if either the cumulative energy change rate or the weighted cumulative energy change rate increases, jumping from a small gradient threshold interval to a large one, it indicates a rapid oscillation divergence. Based on the pre-established gradient threshold-cumulative threshold mapping table, from the next energy cumulative measurement moment, the energy cumulative analysis threshold can be reduced from the previous value according to the corresponding gradient threshold interval, thereby accelerating the analysis speed.
[0077] The method for screening subsynchronous oscillation curves described in this embodiment employs a classifier trained on oscillation data, which can quickly classify oscillation data into two categories: those with oscillation behavior and those without. Furthermore, for data with oscillation behavior, noise is filtered out based on the amplitude of the acquired interharmonic components, and dangerous oscillation behaviors with sensitive frequencies, high energy accumulation, and rapid oscillation divergence are promptly addressed, thereby greatly improving the efficiency and accuracy of screening subsynchronous oscillation data.
[0078] Example 2, Exemplary System
[0079] Figure 2 is a schematic diagram of the structure of a system for screening subsynchronous oscillation curves according to some embodiments. As shown in Figure 2, the system 200 for screening subsynchronous oscillation curves of this disclosure includes:
[0080] The data acquisition module 201 is used to acquire a sequence of amplitude data points of electromagnetic transient simulation phasors to be determined for multiple consecutive time windows according to a custom time window length.
[0081] The data classification module 202 is used to input the sequence of amplitude data points to be determined into a pre-established classifier and output the category of the sequence of amplitude data points to be determined. The category of the sequence of amplitude data points to be determined includes oscillating behavior and no oscillating behavior.
[0082] Data analysis module 203 is used to analyze the sequence of amplitude data points with oscillating behavior to be determined using the time-frequency domain wavelet packet decomposition method, and to obtain the frequency and amplitude of the interharmonic components to be screened.
[0083] The noise filtering module 204 is used to compare the amplitude percentage of the interharmonic component to be screened with a predetermined amplitude threshold, and to take the interharmonic component to be screened after noise filtering as the remaining interharmonic component. The amplitude percentage data refers to the percentage of the amplitude of the interharmonic component to the fundamental frequency.
[0084] The result output module 205 is used to determine the output subsynchronous oscillation curve based on the frequency and amplitude percentage of the remaining interharmonic components.
[0085] In some embodiments, the system further includes a classifier module for establishing a classifier for classifying the sequence of amplitude data points to be determined for electromagnetic transient simulation phasors, specifically:
[0086] Based on a custom time window length, obtain a sequence of historical amplitude data points for electromagnetic transient simulation phasors across multiple consecutive time windows;
[0087] According to the custom dimension number N, the discrete wavelet transform method is used to represent the historical amplitude data point sequence of the electromagnetic transient simulation phasor in each time window as historical coordinate points in N-dimensional space. The number of decomposition layers of the discrete wavelet transform method is equal to the custom dimension number N.
[0088] The categories of historical coordinate points in the N-dimensional space are determined by a preset coordinate point category criterion, and corresponding category labels are assigned according to their categories. The categories of historical coordinate points include those with oscillation behavior and those without oscillation behavior.
[0089] The historical coordinates and their corresponding category labels in the N-dimensional space are used as training data, and a support vector machine is used to train and generate the classifier.
[0090] In some embodiments, the classifier module, according to a custom dimension N, uses the discrete wavelet transform method to represent the historical amplitude data point sequence of the electromagnetic transient simulation phasor for each of multiple time windows as historical coordinate points in N-dimensional space, including:
[0091] The discrete wavelet transform method is used to decompose and reconstruct the historical amplitude data point sequence of the electromagnetic transient simulation phasor for each time window across multiple time windows, thereby determining the wavelet decomposition coefficients d of the nth level corresponding to the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window. in , 1≤i≤I, where I is the total number of time windows, 1≤n≤N;
[0092] According to the wavelet decomposition coefficients d in Generate the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window across multiple time windows, corresponding to the historical coordinate points (d) in N-dimensional space. i1 ,d i2 ,…,d in ,…,d iN ).
[0093] In some embodiments, the coordinate point category criterion is:
[0094] When the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window of multiple time windows corresponds to the historical coordinate point in the N-dimensional space (d... i1 ,d i2 ,…,d in ,…,d iN When ), if d ia With d ib When the absolute value of all differences is less than d0, the historical coordinate point is classified as having no oscillation behavior; otherwise, in d... ia With d ib If the absolute value of the difference is greater than or equal to d0, the category of the historical coordinate point is oscillating behavior, where d0 is a positive number less than 1, 1≤i≤I, 1≤a, b≤N, and a≠b.
[0095] In some embodiments, the system further includes an amplitude threshold module for determining an amplitude threshold for filtering noise in a window exhibiting oscillatory behavior, specifically:
[0096] The time-frequency domain wavelet packet decomposition method is used to analyze the sequence of historical amplitude data points corresponding to historical coordinate points with oscillating behavior, obtain the amplitude of the corresponding historical interharmonic components, and calculate the amplitude percentage of the historical interharmonic components.
[0097] According to the set interharmonic component classification rules, the category of the historical interharmonic component is determined based on the amplitude percentage of the historical interharmonic component, and a corresponding category label is added. The categories of the historical interharmonic components include those that need attention and those that do not need attention.
[0098] Based on the amplitude percentage and category label of the historical interharmonic components, a support vector machine learner is trained to obtain the boundary point on a one-dimensional number axis for the amplitude percentage data of the classes that need attention and the classes that do not need attention. The boundary point is the determined amplitude threshold.
[0099] In some embodiments, the interharmonic component classification rules applied in the amplitude threshold module include:
[0100] Historical interharmonic components with an amplitude percentage below L are defined as the no-concern category;
[0101] Historical interharmonic components with an amplitude percentage higher than M% are defined as the class requiring attention, L. <M。
[0102] For historical interharmonic components that do not meet the above two definitions, if their amplitude percentage is higher than K times the average amplitude percentage of all historical interharmonic components in the same time window, the category is defined as a class that needs attention; otherwise, the category is defined as a class that does not need attention, where K is a natural number.
[0103] In some embodiments, the wavelet mother function used in the discrete wavelet transform method employed in the classifier module or the time-frequency domain wavelet packet decomposition method employed in the data analysis module and the amplitude threshold module is a third-order Daubechies series wavelet, whose expression is: H(ω)=0.0249086-0.0604169e -iω -0.095467e -2iω +0.325186e -3iω
[0104] ω is the amplitude point sequence of the input electromagnetic transient simulation phasor with oscillating behavior, and H(ω) is the amplitude and frequency of the interharmonic components obtained after hierarchical decomposition of the input.
[0105] In some embodiments, the noise filtering module 204 compares the amplitude percentage of the interharmonic components to be filtered with a predetermined amplitude threshold, and uses the filtered interharmonic components as the remaining interharmonic components, including:
[0106] The amplitude percentage of the interharmonic components to be screened is compared with a predetermined amplitude threshold, and the interharmonic components to be screened whose amplitude percentage is less than the amplitude threshold are filtered out as noise.
[0107] The remaining part of the interharmonic components to be screened after filtering out noise is taken as the remaining interharmonic components.
[0108] In some embodiments, the result output module 205 determines the output subsynchronous oscillation curve based on the frequency and amplitude percentage of the remaining interharmonic components, including:
[0109] S1. Based on the frequency of each interharmonic component in the sensitive frequency band and the remaining interharmonic components, determine the energy accumulation benchmark threshold and energy accumulation weight value of each interharmonic component. The sensitive frequency band is a frequency range composed of the upper limit and lower limit values obtained by adding and subtracting a custom frequency step size from the difference between the power frequency and the motor modal frequency during electromagnetic transient simulation. Energy accumulation refers to the percentage of amplitude of the interharmonic component integrated over time. The first energy accumulation benchmark threshold of the interharmonic component with frequency in the sensitive frequency band is less than the second energy accumulation benchmark threshold of the interharmonic component with frequency not in the sensitive frequency band. The first energy accumulation weight value of the interharmonic component with frequency in the sensitive frequency band is greater than the second energy accumulation weight value of the interharmonic component with frequency not in the sensitive frequency band. The sum of the first energy accumulation weight value and the second energy accumulation weight value is between the frequency step size and twice the frequency step size. The first energy accumulation weight value is k times the second energy accumulation weight value, where k is a positive number not less than 3.
[0110] S2. According to the set time interval, calculate the cumulative energy value of each interharmonic component in the j-th time interval, and calculate the cumulative weighted sum of energy in the j-th time interval based on the cumulative energy weight value of each interharmonic component and the cumulative energy value in the j-th time interval. j is a natural number and the initial value of j is 1.
[0111] S3. When the cumulative energy value of each interharmonic component in the j-th time interval is less than its corresponding cumulative energy analysis threshold, and the cumulative weighted sum of energy in the j-th time interval is less than the set cumulative weighted sum of energy analysis threshold, execute S4; otherwise, execute S7. The cumulative weighted sum of energy analysis threshold is any positive integer between [5, 50]. When j = 1, the cumulative energy analysis threshold is equal to the cumulative energy reference threshold, and the cumulative energy value and the cumulative weighted sum of energy in the (j-1)-th time interval are equal to the cumulative energy value and the cumulative weighted sum of energy in the (j-1)-th time interval, respectively.
[0112] S4. Calculate the cumulative energy change rate of each interharmonic component based on the cumulative energy value of each interharmonic component in the j-th time interval and the (j-1)-th time interval, and calculate the weighted cumulative energy change rate based on the weighted sum of the cumulative energy values in the j-th time interval and the (j-1)-th time interval.
[0113] S5. Based on the set rate of change gradient threshold mapping relationship table, determine the first gradient threshold interval corresponding to the j-th cumulative energy change rate of each interharmonic component, and the second gradient threshold interval corresponding to the j-th cumulative weighted energy change rate. The first gradient threshold interval is generated by setting P1 gradient thresholds in ascending order of energy cumulative change rate, and making two adjacent gradient thresholds form a gradient threshold interval, resulting in P1-1 value intervals arranged in ascending order. The second gradient threshold interval is generated by setting P2 gradient thresholds in ascending order of energy cumulative weighted change rate, and making two adjacent gradient thresholds form a gradient threshold interval, resulting in P2-1 value intervals arranged in ascending order.
[0114] S6. When the first gradient threshold interval of the j-th cumulative energy change rate is greater than the first gradient threshold interval of the (j-1)-th cumulative energy change rate, or when the second gradient threshold interval corresponding to the j-th cumulative weighted energy change rate is greater than the second gradient threshold interval corresponding to the (j-1)-th cumulative energy change rate, according to the set gradient threshold energy cumulative threshold mapping relationship table, the energy cumulative threshold corresponding to the first gradient threshold interval of the (j-1)-th cumulative energy change rate is used as the energy cumulative analysis threshold for the (j+1)-th time interval. Otherwise, the energy cumulative threshold corresponding to the first gradient threshold interval of the j-th cumulative energy change rate is used as the energy cumulative analysis threshold for the (j+1)-th time interval. Let j = j + 1, and execute S2. When j = 1, the first gradient threshold interval of the (j-1)-th cumulative energy change rate is equal to the first gradient threshold interval of the j-th cumulative energy change rate.
[0115] S7. Output the interharmonic components from the first time interval to the j-th time interval as the selected sub-frequency oscillation curves.
[0116] The system for screening subsynchronous oscillation curves described in this embodiment has the same steps as the method for screening subsynchronous oscillation curves, and achieves the same technical effect. Therefore, this embodiment will not be described in detail here.
[0117] Example 3: Exemplary Electronic Device
[0118] Figure 3 is a schematic diagram of an electronic device according to some embodiments. The electronic device can be either or both of a first device and a second device, or a standalone device independent of them, which can communicate with the first and second devices to receive acquired input signals from them. Figure 3 shows a block diagram of an electronic device according to an embodiment of the present disclosure. As shown in Figure 3, the electronic device includes one or more processors 301 and a memory 302.
[0119] The processor 301 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.
[0120] The memory 302 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 301 may execute the program instructions to implement the methods of the various embodiments disclosed above and / or other desired functions. In one example, the electronic device may also include an input device 303 and an output device 304, these components being interconnected via a bus system and / or other forms of connection mechanisms (not shown).
[0121] In addition, the input device 303 may also include, for example, a keyboard, a mouse, etc.
[0122] The output device 304 can output various information to the outside. The output device 304 may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.
[0123] Of course, for simplicity, Figure 3 only shows some of the components of the electronic device that are relevant to this disclosure, omitting components such as buses, input / output interfaces, etc. In addition, the electronic device may include any other suitable components depending on the application.
[0124] Example 4: Exemplary Computer Program Product and Computer-Readable Storage Medium
[0125] In addition to the methods and apparatus described above, embodiments of this disclosure also provide a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the methods for screening subsynchronous oscillation curves according to various embodiments of this disclosure as described in the "Exemplary Methods" section above.
[0126] The computer program product can be written in any combination of one or more programming languages to perform the operations of the embodiments of this disclosure. The programming languages include object-oriented programming languages such as Java and C++, as well as conventional procedural programming languages such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0127] Furthermore, embodiments of this disclosure also provide a computer-readable storage medium (e.g., a non-transitory computer-readable storage medium) having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform steps in the methods for screening subsynchronous oscillation curves according to various embodiments of this disclosure as described in the "Exemplary Methods" section above.
[0128] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More detailed examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.
[0129] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the detailed information disclosed above is for illustrative and facilitative purposes only, and is not a limitation. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned detailed information.
[0130] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.
[0131] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.
[0132] The apparatus and methods of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.
[0133] It should also be noted that in the apparatus, devices, and methods of this disclosure, the components or steps are decomposable and / or recombinable. Such decomposition and / or recombination should be considered equivalent to the present disclosure. The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.
[0134] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.
Claims
1. A method for screening subsynchronous oscillation curves, comprising: Based on the custom time window length, obtain the sequence of amplitude data points of electromagnetic transient simulation phasors to be determined for multiple consecutive time windows; The sequence of amplitude data points to be determined is input into a pre-established classifier, which outputs the category of the sequence of amplitude data points to be determined. The category of the sequence of amplitude data points to be determined includes those with oscillating behavior and those without oscillating behavior. The time-frequency domain wavelet packet decomposition method is used to analyze the sequence of amplitude data points with oscillating behavior to obtain the frequency and amplitude of the interharmonic components to be screened. The amplitude percentage of the interharmonic component to be screened is compared with a predetermined amplitude threshold, and the interharmonic component to be screened after noise is filtered out is taken as the remaining interharmonic component. The amplitude percentage refers to the percentage of the amplitude of the interharmonic component relative to the fundamental frequency. The output subsynchronous oscillation curve is determined based on the frequency of the remaining interharmonic components and the amplitude percentage.
2. The method according to claim 1, wherein, Before obtaining the amplitude data point sequence of the electromagnetic transient simulation phasors for the multiple consecutive time windows according to the custom time window length, a classifier is established to classify the amplitude data point sequence of the electromagnetic transient simulation phasors to be determined. The establishment of the classifier for classifying the amplitude data point sequence of the electromagnetic transient simulation phasors to be determined includes: Based on a custom time window length, obtain a sequence of historical amplitude data points for electromagnetic transient simulation phasors across multiple consecutive time windows; According to the custom dimension number N, the discrete wavelet transform method is used to represent the historical amplitude data point sequence of the electromagnetic transient simulation phasor in each of the multiple time windows as historical coordinate points in N-dimensional space. The number of decomposition layers of the discrete wavelet transform method is equal to the custom dimension number N. The categories of historical coordinate points in the N-dimensional space are determined by using a preset coordinate point category criterion, and corresponding category labels are assigned according to the categories of the historical coordinate points. The categories of the historical coordinate points include those with oscillation behavior and those without oscillation behavior. The classifier is generated by training the historical coordinate points in the N-dimensional space and the corresponding category labels of the historical coordinate points as training data using a support vector machine.
3. The method according to claim 2, wherein, The step of representing the historical amplitude data point sequence of the electromagnetic transient simulation phasor in each of the multiple time windows as historical coordinate points in N-dimensional space using the discrete wavelet transform method, according to the custom dimension N, includes: The discrete wavelet transform method is used to decompose and reconstruct the historical amplitude data point sequence of the electromagnetic transient simulation phasor in each of the multiple time windows, and the wavelet decomposition coefficients d of the nth level corresponding to the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window are determined. in Where 1≤i≤I, I is the total number of time windows, and 1≤n≤N; According to the wavelet decomposition coefficients d in Generate the historical coordinate points (d) in N-dimensional space corresponding to the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window of the plurality of time windows. i1 ,d i2 ,…,d in ,…,d iN ).
4. The method according to claim 2, wherein, The preset coordinate point category criterion is: When the historical coordinate point in the N-dimensional space corresponding to the historical amplitude data point sequence of the electromagnetic transient simulation phasor in the i-th time window of the plurality of time windows is (d i1 ,d i2 ,…,d in ,…,d iN When ), at d ia With d ib If the absolute value of all differences is less than d0, the historical coordinate point is classified as having no oscillation behavior; otherwise, in d... ia With d ib If the absolute value of the difference is greater than or equal to d0, the historical coordinate point is classified as having oscillating behavior, where d0 is a positive number less than 1, 1≤i≤1, 1≤a, b≤N, and a≠b.
5. The method according to claim 2, wherein, After determining the category of historical coordinate points in the N-dimensional space using the preset coordinate point category criterion, the process includes determining an amplitude threshold for filtering noise from windows exhibiting oscillating behavior. Determining the amplitude threshold for filtering noise from windows exhibiting oscillating behavior includes: The time-frequency domain wavelet packet decomposition method is used to analyze the sequence of historical amplitude data points corresponding to historical coordinate points with oscillating behavior, obtain the amplitude of the corresponding historical interharmonic components, and calculate the amplitude percentage of the historical interharmonic components. According to the set interharmonic component classification rules, the category of the historical interharmonic component is determined based on the amplitude percentage of the historical interharmonic component, and a corresponding category label is added. The categories of the historical interharmonic components include those requiring attention and those not requiring attention. Based on the amplitude percentage of the historical interharmonic components and the category label, a support vector machine learner is trained to obtain the boundary point on a one-dimensional number axis for the amplitude percentage data of the class that needs attention and the class that does not need attention. The boundary point is the determined amplitude threshold.
6. The method according to claim 5, wherein, The interharmonic component classification rules include: Historical interharmonic components with an amplitude percentage below L are defined as the no-concern category; Historical interharmonic components with an amplitude percentage higher than M% are defined as the class requiring attention, where L <M; For historical interharmonic components that do not meet the above two definitions, if their amplitude percentage is higher than K times the average amplitude percentage of all historical interharmonic components in the same time window, the category is defined as a class requiring attention; otherwise, the category is defined as a class not requiring attention, where K is a natural number.
7. The method according to claim 1, 2 or 5, wherein, The wavelet mother function used in the discrete wavelet transform method or the time-frequency domain wavelet packet decomposition method is a third-order Daubechies series wavelet, whose expression is: H(ω)=0.0249086-0.0604169e -iω -0.095467e -2iω +0.325186e -3iω Where ω is the amplitude point sequence of the input electromagnetic transient simulation phasor with oscillating behavior, and H(ω) is the amplitude and frequency of the interharmonic components obtained after hierarchical decomposition of the input.
8. The method according to claim 1, wherein, The step of comparing the amplitude percentage of the interharmonic components to be screened with a predetermined amplitude threshold, and taking the noise-filtered interharmonic components as the remaining interharmonic components, includes: The amplitude percentage of the interharmonic components to be screened is compared with a predetermined amplitude threshold, and the interharmonic components to be screened whose amplitude percentage is less than the amplitude threshold are filtered out as noise. The remaining part of the interharmonic components to be screened after filtering out noise is taken as the remaining interharmonic components.
9. The method according to claim 1, wherein, The determination of the output subsynchronous oscillation curve based on the frequency and amplitude percentage of the remaining interharmonic components includes: S1. Based on the sensitive frequency band and the frequency of each interharmonic component in the remaining interharmonic components, determine the energy accumulation reference threshold and energy accumulation weight value of each interharmonic component. The sensitive frequency band is a frequency range composed of an upper limit and a lower limit obtained by adding and subtracting a custom frequency step size from the difference between the power frequency and the motor modal frequency during electromagnetic transient simulation. Energy accumulation refers to the percentage of amplitude of the interharmonic component integrated over time. The first energy accumulation reference threshold of the interharmonic component with frequency in the sensitive frequency band is less than the second energy accumulation reference threshold of the interharmonic component with frequency not in the sensitive frequency band. The first energy accumulation weight value of the interharmonic component with frequency in the sensitive frequency band is greater than the second energy accumulation weight value of the interharmonic component with frequency not in the sensitive frequency band. The sum of the first energy accumulation weight value and the second energy accumulation weight value is between the frequency step size and twice the frequency step size. The first energy accumulation weight value is k times the second energy accumulation weight value, where k is a positive number not less than 3. S2. According to the set time interval, calculate the cumulative energy value of each interharmonic component in the j-th time interval, and calculate the cumulative weighted sum of energy in the j-th time interval based on the cumulative energy weight value of each interharmonic component and the cumulative energy value in the j-th time interval, where j is a natural number and the initial value of j is 1; S3. When the cumulative energy value of each interharmonic component in the j-th time interval is less than its corresponding cumulative energy analysis threshold, and the cumulative weighted sum of energy in the j-th time interval is less than the set cumulative weighted sum of energy analysis threshold, execute S4; otherwise, execute S7. Wherein, the cumulative weighted sum of energy analysis threshold is any positive integer between [5, 50]. When j = 1, the cumulative energy analysis threshold is equal to the cumulative energy reference threshold, and the cumulative energy value and the cumulative weighted sum of energy in the (j-1)-th time interval are equal to the cumulative energy value and the cumulative weighted sum of energy in the (j-1)-th time interval, respectively. S4. Calculate the j-th cumulative energy change rate of each interharmonic component based on the cumulative energy value of each interharmonic component in the j-th time interval and the (j-1)-th time interval, and calculate the j-th cumulative weighted change rate of energy based on the weighted sum of the cumulative energy values in the j-th time interval and the (j-1)-th time interval; S5. Based on the set rate of change gradient threshold mapping relationship table, determine the first gradient threshold interval corresponding to the j-th cumulative energy change rate of each interharmonic component, and the second gradient threshold interval corresponding to the j-th cumulative weighted energy change rate. The first gradient threshold interval is a P1-1 value interval generated by setting P1 gradient thresholds in ascending order of energy cumulative change rate and making two adjacent gradient thresholds form a gradient threshold interval. The second gradient threshold interval is a P2-1 value interval generated by setting P2 gradient thresholds in ascending order of energy cumulative weighted change rate and making two adjacent gradient thresholds form a gradient threshold interval. S6. When the first gradient threshold interval of the j-th cumulative energy change rate is greater than the first gradient threshold interval of the (j-1)-th cumulative energy change rate, or when the second gradient threshold interval corresponding to the j-th cumulative weighted energy change rate is greater than the second gradient threshold interval corresponding to the (j-1)-th cumulative energy change rate, according to the set gradient threshold energy cumulative threshold mapping relationship table, the energy cumulative threshold corresponding to the first gradient threshold interval of the (j-1)-th cumulative energy change rate is used as the energy cumulative analysis threshold for the (j+1)-th time interval. Otherwise, the energy cumulative threshold corresponding to the first gradient threshold interval of the j-th cumulative energy change rate is used as the energy cumulative analysis threshold for the (j+1)-th time interval. Let j = j + 1, and execute S2. Where, when j = 1, the first gradient threshold interval of the (j-1)-th cumulative energy change rate is equal to the first gradient threshold interval of the j-th cumulative energy change rate. S7. Output the interharmonic components from the first time interval to the j-th time interval as the selected sub-frequency oscillation curves.
10. A system for screening subsynchronous oscillation curves, comprising: The system includes a data acquisition module, a data classification module, a data analysis module, a noise filtering module, and a result output module. The data acquisition module is used to acquire a sequence of amplitude data points of electromagnetic transient simulation phasors to be determined for multiple consecutive time windows, based on a custom time window length. The data classification module is used to input the sequence of amplitude data points to be determined into a pre-established classifier and output the category of the sequence of amplitude data points to be determined. The category of the sequence of amplitude data points to be determined includes oscillating behavior and no oscillating behavior. The data analysis module is used to analyze the sequence of amplitude data points with oscillating behavior to be determined using the time-frequency domain wavelet packet decomposition method, and to obtain the frequency and amplitude of the interharmonic components to be screened. The noise filtering module is used to compare the amplitude percentage of the interharmonic component to be screened with a predetermined amplitude threshold, and to take the interharmonic component to be screened after noise filtering as the remaining interharmonic component. The amplitude percentage data refers to the percentage of the amplitude of the interharmonic component to the fundamental frequency. The result output module is used to determine the output subsynchronous oscillation curve based on the frequency and amplitude percentage of the remaining interharmonic components.
11. A computer-readable storage medium, wherein, The computer-readable storage medium stores a computer program for performing the method according to any one of claims 1 to 9.
12. An electronic device, comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method according to any one of claims 1 to 9.