Subsynchronous oscillation identification method based on VMD-MSST-HHT and identification system thereof

Abstract

The invention provides a subsynchronous oscillation identification method based on VMD-MSST-HHT and an identification system thereof, which are used for realizing accurate identification of subsynchronous oscillation signals. The invention relates to the technical field of new energy power system signal identification. The method comprises the following steps: decomposing an acquired original signal into K modal components by adopting variational mode decomposition (VMD); performing multiple synchronous compression transformation (MSST) on each modal component decomposed by the VMD to obtain MSST time-frequency distribution of each modal, and then performing superposition; and carrying out Hilbert-Huang transform HHT on the superposed result to extract the instantaneous characteristics of the subsynchronous oscillation signal, thereby completing the identification of the subsynchronous oscillation signal. The subsynchronous oscillation signal identification method is suitable for subsynchronous oscillation signal identification of a new energy power system, and the identification method has the advantages of being high in modal separation capacity, high in time-frequency resolution, good in anti-noise performance and the like.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy power system signal identification, and particularly relates to a subsynchronous oscillation identification method based on VMD-MSST-HHT. BACKGROUND

[0002] With the increasing penetration of new energy power electronic devices in power systems, power electronicization has become a development trend of power systems. The dynamic interaction characteristics of the system are complex, and a large number of new energy units connected to the grid through converters interact with the grid, which aggravates the risk of subsynchronous oscillation of the system. Subsynchronous oscillation accidents frequently occur in power systems, which poses a serious threat to the safe and stable operation of power systems.

[0003] Therefore, fast and accurate identification of subsynchronous oscillation signals can provide a reference for taking effective suppression measures in the early stage of oscillation, which is of great significance to avoid further expansion of accidents and ensure the safe and stable operation of power systems. At present, the analysis of subsynchronous oscillation in power grids mainly relies on the data collected by the phasor measurement unit (PMU) device. In engineering applications, fast Fourier transform (FFT), wavelet transform (WT) and Hilbert-Huang transform (HHT) are mainly used to analyze subsynchronous oscillation of the recording wave signal. However, there are a series of problems, such as modal aliasing, spectrum leakage and fence effect of FFT, which will lead to incorrect analysis results. WT has the limitation of fixed basis function, and the extraction effect of different basis functions on signal characteristics is significantly different. At the same time, the time window width and frequency resolution of WT are mutually restricted, and it is difficult to achieve time accuracy in high frequency band and frequency accuracy in low frequency band at the same time. HHT has boundary effect, and false frequency components will be generated at both ends of the signal due to data truncation, which affects the analysis of transient process. At the same time, HHT is extremely sensitive to noise, and it is also difficult to process signals with rapid amplitude-frequency changes.

[0004] The current mainstream research direction is to pre-process the oscillation signal to improve the accuracy of identification, such as synchronous compression transform SST, empirical mode decomposition EMD, multiple synchronous compression transform MSST, variational mode decomposition VMD, synchronous extraction transform SET and the like. However, although SST can improve the time-frequency concentration, it is still difficult to completely separate if the signal component frequencies are adjacent. When EMD is decomposed, different frequency components may be mixed in the same IMF, so it has the problem of mode aliasing, and EMD is also limited by the end effect because of the lack of data reference at both ends of the signal, resulting in false oscillation components, affecting the analysis of the transient process. MSST is sensitive to aliasing signals, and HHT depends on the compression number of MSST, which may cause saturation. VMD has poor adaptability to non-stationary abrupt components, and the analysis accuracy of signals with rapid amplitude-frequency changes (such as time-varying subsynchronous oscillation) is insufficient, making it difficult to be applied to new energy power systems. SET extracts instantaneous frequency through phase information, and noise will interfere with phase estimation, resulting in false energy or broken ridge lines in the time-frequency diagram. SUMMARY

[0005] The purpose of the present application is to provide a VMD-MSST-HHT-based subsynchronous oscillation identification method, which is used to solve the technical problems mentioned in the background art and realize accurate identification of subsynchronous oscillation signals.

[0006] To achieve the above purpose, the present application provides the following technical solutions: The present application provides a VMD-MSST-HHT-based subsynchronous oscillation identification method, which comprises: using variational mode decomposition VMD to decompose the collected original signal into K modal components ; performing multiple synchronous compression transform MSST on each modal component decomposed by VMD, obtaining the MSST time-frequency distribution of each mode, and then superimposing; performing Hilbert-Huang transform HHT on the superimposed result to extract the instantaneous characteristics of the subsynchronous oscillation signal, and completing the identification of the subsynchronous oscillation signal.

[0007] Further, the above original signal is a multimodal oscillation signal, and the original signal is collected by a phasor measurement unit.

[0008] Further, the above original signal is decomposed into K modal components , specifically: the original signal is adaptively decomposed intoK modal components each modal component has an instantaneous frequency ; each modal component is analyzed to obtain an analytic signal of each modal component; the analytic signal is multiplied by the estimated center frequency to shift the single side frequency spectrum to the corresponding baseband; the bandwidth of each mode is estimated by using a Gaussian smoothing method, and a mathematical model of the VMD is established; by constraint variation conversion, parameter iterative updating and optimal solution solving on the established mathematical model, the K modal components IMF are obtained as the decomposition results of the VMD.

[0009] Further, each modal component is subjected to a multiple synchronous compression transform MSST, which is specifically: the original signal subjected to the VMD decomposition is subjected to a synchronous compression transform SST to obtain a short-time Fourier transform time-frequency spectrum ; the is subjected to a synchronous compression transform to obtain an SST expression; the SST expression is subjected to multiple iterative synchronous compression transforms, and the number of iterations is N, to obtain the MSST time-frequency distribution of each mode.

[0010] Further, the instantaneous characteristics include the instantaneous amplitude, the instantaneous frequency, the instantaneous phase and the attenuation factor of the subsynchronous oscillation.

[0011] The subsynchronous oscillation identification method based on the VMD-MSST-HHT can be realized by using computer software, and accordingly, the present application further provides a subsynchronous oscillation identification system based on the VMD-MSST-HHT, which comprises: a storage device for decomposing the collected original signal into K modal components by using the VMD; a storage device for performing a multiple synchronous compression transform MSST on each modal component subjected to the VMD decomposition to obtain the MSST time-frequency distribution of each mode, and then performing superposition; a storage device for performing a Hilbert-Huang transform HHT on the superimposed result to extract the instantaneous characteristics of the subsynchronous oscillation signal, thereby completing the subsynchronous oscillation signal identification.

[0012] Further, the storage device for decomposing the collected original signal into K modal components by using the variational mode decomposition (VMD) is used to decompose the original signal into K modal components. K Specifically, the original signal is adaptively decomposed into K modal components, each of which has an instantaneous frequency. K The analytic signal of each modal component is obtained. The analytic signal is multiplied by the estimated center frequency to shift the single side frequency spectrum to the corresponding baseband. The bandwidth of each modal component is estimated by using the Gaussian smoothing method, and a mathematical model of the variational constraint of VMD is established. The mathematical model is transformed by constraint variation, the parameters are iteratively updated, and the optimal solution is solved, so that K modal components (IMFs) are obtained as the decomposition results of VMD. K

[0013] The application further provides a computer readable storage medium, and the computer readable storage medium stores a computer program.

[0014] The application further provides a computer device, which comprises a memory and a processor.

[0015] The application has the following beneficial effects: 1. The application provides a VMD-MSST-HHT-based sub-synchronous oscillation identification method, which can effectively improve the modal separation capability, realize high resolution, process complex multi-component, strong noise or modal aliasing signals, and accurately identify the sub-synchronous oscillation signal parameters.

[0016] ​​​​​​​​​​​Further, the present application can effectively improve the modal separation ability by carrying out VMD and MSST pretreatment before HHT, and still accurately identify the parameters of the subsynchronous oscillation signal when dealing with multi-component, strong noise and modal aliasing signals.

[0017] Further, the present application adopts VMD to separate signal components in advance, avoids the modal aliasing problem of MSST and HHT, and also adopts MSST to improve energy aggregation through multiple compression to adapt to the analysis of adjacent frequency components.

[0018] 2、The present application is suitable for subsynchronous oscillation signal identification of new energy power system, and the identification method has the characteristics of strong modal separation ability, high time-frequency resolution and good noise resistance. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The flow chart of the VMD-MSST-HHT-based subsynchronous oscillation identification method of the present application; Figure 2 The schematic diagram of the subsynchronous oscillation identification test signal of the present application; Figure 3 The time-frequency ridge effect schematic diagram of the subsynchronous oscillation identification of the traditional method and the method of the present application, wherein figure (a) is the time-frequency ridge effect schematic diagram of the traditional STFT method, figure (b) is the time-frequency ridge effect schematic diagram of the traditional SST method, figure (c) is the time-frequency ridge effect schematic diagram of the traditional SET method, and figure (d) is the time-frequency ridge effect schematic diagram of the VMD-MSST method of the present application; Figure 4 The effect schematic diagram of the subsynchronous oscillation test signal decomposition of the traditional SST method; Figure 5 The effect schematic diagram of the subsynchronous oscillation test signal decomposition of the traditional SET method; Figure 6 The effect schematic diagram of the subsynchronous oscillation test signal decomposition of the traditional VMD method; Figure 7 The effect schematic diagram of the subsynchronous oscillation test signal decomposition of the method of the present application. DETAILED DESCRIPTION

[0020] In the following description, the specific implementation details (such as operation flow, data processing steps and example parameters) of the "VMD-MSST-HHT-based subsynchronous oscillation identification method" provided by the specification are fundamentally intended to be illustratively described rather than limitatively defined, and are intended to help those skilled in the art to thoroughly understand the principles and implementation of the present application; however, those skilled in the art should clearly understand that these details only represent one of the possible embodiments, and the core concept of the present application can be realized by other technical means or alternative schemes without departing from the spirit thereof, and the omission of details of conventional experimental methods and devices known in the art in the specification is to avoid redundant information interfering with the understanding of the innovative points, which does not mean that these known technologies are not required in implementation. Technical personnel should be able to supplement and use them based on their professional knowledge.

[0021] The specific embodiments of the present application will be further described in detail below with reference to the accompanying drawings. The following embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any form. It should be noted that, for those skilled in the art, without departing from the concept of the present application, a number of changes and improvements can be made, which are within the scope of protection of the present application.

[0022] Embodiment one, combined Figure 1 and Figure 2 This embodiment provides a VMD-MSST-HHT-based subsynchronous oscillation identification method. The flow of the identification method is shown in Figure 1 , which shows the steps of mode decomposition, multiple synchronous compression and parameter identification of the method for subsynchronous oscillation signals.

[0023] The identification method comprises the following steps: Step S1: using variational mode decomposition VMD to decompose the collected original signal into K modal components ; Step S2: performing multiple synchronous compression transform MSST on each modal component decomposed by VMD, and then superimposing after obtaining the MSST time-frequency distribution of each mode; Step S3: performing Hilbert-Huang transform HHT on the superimposed result to extract the instantaneous characteristics of the subsynchronous oscillation signal, and completing the identification of the subsynchronous oscillation signal.

[0024] In actual application, when subsynchronous oscillation occurs in the power system, the data collected in the phasor measurement unit (PMU) device in the power grid is identified, and the test signal is as follows Figure 2shown; first, the original signal recorded in the PMU is adaptively decomposed into several modal components by using the VMD (Variational Mode Decomposition) K The center frequency and bandwidth of each mode are iteratively optimized by a variational framework, realizing the frequency domain adaptive decomposition of the signal; then, each mode decomposed by VMD is subjected to MSST (Multiple Synchronous Synchronous Transform), eliminating the time-frequency energy dispersion and improving the clarity and energy concentration of the time-frequency representation of the non-stationary signal, so as to finally obtain the MSST time-frequency distribution of each mode; finally, each mode subjected to MSST is subjected to Hilbert transform, extracting the instantaneous features and identifying the instantaneous amplitude, instantaneous frequency, instantaneous phase and attenuation factor of the subsynchronous oscillation.

[0025] Embodiment II, the embodiment is a specific description of the subsynchronous oscillation identification method based on VMD-MSST-HHT proposed in the above embodiment I; Step S1: the original signal collected is adaptively decomposed into several modal components by using the VMD (Variational Mode Decomposition) K Specifically, A set of commonly used ideal multi-modal oscillation signals are constructed as original signals, which can be expressed as:

[0026] wherein, represents the initial amplitude, and the subscript k represents the K modal signal component, represents the signal attenuation term, represents the attenuation factor, represents the frequency of the signal, represents the moment, represents the initial phase of the signal.

[0027] The original signal is adaptively decomposed into several modal components K , each of which has an instantaneous frequency , and is expressed as:

[0028] wherein, represents the instantaneous amplitude, and , represents the phase at any moment, and .

[0029] ​​​​​​​​Analyze each modal component to obtain the analytical signal of each modal component, which can be expressed as:

[0030] in, is the Dirac distribution function, represents the convolution signal, represents an imaginary number, Indicates the K modal signal components.

[0031] By multiplying the analytical signal by the estimated center frequency, the single-sided spectrum is shifted to the corresponding baseband, expressed as:

[0032] in, is the phasor description of the center frequency on the complex plane.

[0033] The bandwidth of each mode is estimated using the Gaussian smoothing method, and then the mathematical model of the variational limit of VMD is established:

[0034] in, is the original input signal, is the modal component of each IMF after decomposition; is the center frequency of each IMF after decomposition. for t The partial derivative operator at the moment; * represents the convolution operator.

[0035] By introducing the quadratic penalty factor and Lagrange multipliers , transforming the constrained variational problem into an unconstrained variational problem:

[0036] Iterative update using alternating direction multiplier method 、 、 , find the saddle point of the Lagrangian function, which is the optimal solution.

[0037]

[0038] in, Respectively represent Finally, we can get K The modal components IMF are the results of VMD.

[0039] The embodiment avoids the mode aliasing problem of the multiple synchronous compression transform (MSST) and the Hilbert-Huang transform (HHT) by adopting the variational mode decomposition (VMD) to pre-separate signal components.

[0040] Step S2: performing the multiple synchronous compression transform (MSST) on each mode component decomposed by the VMD to obtain the MSST time-frequency distribution of each mode; After the MSST time-frequency distribution of each mode is obtained, superposition is performed. Specifically, The original signal processed by the VMD is subjected to the synchronous compression transform (SST), and the SST expression is obtained. The short-time Fourier transform time-frequency spectrum is obtained.

[0041] wherein, represents the instantaneous amplitude, represents the signal phase, is a window function, represents a signal, represents the Fourier transform of is According to the expression after the Taylor series expansion, represents the differential.

[0042] The SST expression is obtained by performing the synchronous compression transform once on .

[0043] wherein, represents an impulse function, is an instantaneous frequency estimation value, is a synchronous compression transform coefficient.

[0044] The MSST expression is obtained by performing the multiple iterative synchronous compression transform on the SST expression, and the number of iterations is N .

[0045] Each mode is subjected to the multiple synchronous compression transform, and the time-frequency concentration is enhanced.

[0046] Step S3: performing the Hilbert-Huang transform (HHT) on the superimposed result to extract the instantaneous features of the subsynchronous oscillation signal and complete the subsynchronous oscillation signal identification.

[0047] Specifically, The single-mode component signal is taken as an example. The single-mode component signal The HHT transform of the signal can be expressed as:

[0048] The analytic signal of the signal can be expressed as:

[0049] wherein, and respectively represent the instantaneous amplitude and the instantaneous phase, and the instantaneous frequency of the single mode component can be obtained through the derivative of the instantaneous phase;

[0050] The single mode component is fitted to obtain the attenuation factor .

[0051] In summary, the VMD-MSST-HHT-based subsynchronous oscillation identification method proposed in the embodiment can effectively improve the modal separation capability, realize high resolution, process complex multi-component, strong noise or modal aliasing signals, and accurately identify the parameters of the subsynchronous oscillation signal. Further, the VMD and MSST preprocessing before the HHT transform can effectively improve the modal separation capability, and the parameters of the subsynchronous oscillation signal can still be accurately identified when dealing with multi-component, strong noise and modal aliasing signals. Further, the VMD is used to separate the signal components in advance to avoid the modal aliasing problem of MSST and HHT; and the MSST is used to improve the energy concentration through multiple compressions to adapt to the analysis of adjacent frequency components.

[0052] Embodiment three, the VMD-MSST-HHT-based subsynchronous oscillation identification method described in any one of the above embodiments can be implemented by computer software, therefore, the embodiment provides a VMD-MSST-HHT-based subsynchronous oscillation identification system, the system comprises: a storage device for decomposing the collected original signal into K modal components by using the VMD; a processor for performing MSST on each modal component A storage device for performing multiple synchronous compression transforms (MSSTs) to obtain the MSST time-frequency distribution of each mode and then performing superposition; A storage device for performing Hilbert-Huang transform (HHT) on the superimposed result to extract the instantaneous characteristics of the subsynchronous oscillation signal and complete the identification of the subsynchronous oscillation signal.

[0053] Furthermore, the above method is used to decompose the original signal collected by using variational mode decomposition VMD. Decompose into K modal components The original signal is stored in the storage device Decomposed into K modal components Specifically: The original signal Adaptive decomposition K modal components , each modal component has an instantaneous frequency ; For each modal component Perform analysis to obtain analytical signals of each modal component; Multiply the analytical signal by the estimated center frequency to shift the single-sided spectrum to the corresponding baseband; The bandwidth of each mode is estimated using Gaussian smoothing method, and a mathematical model of variational limitation of VMD is established; By performing constrained variational transformation, iterative parameter updating and optimal solution on the established mathematical model, we can obtain K The modal components IMF are used as the decomposition results of VMD.

[0054] Implementation method 4: This implementation method provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the subsynchronous oscillation identification method based on VMD-MSST-HHT described in any one of the above implementation methods is executed.

[0055] Implementation method five. This implementation method provides a computer device, which includes a memory and a processor, wherein a computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes a subsynchronous oscillation identification method based on VMD-MSST-HHT as described in any one of the above implementation methods.

[0056] The computer device provided by the embodiment includes a hardware device of a general type, which is not represented in the form of a diagram. The system includes a processor and a memory, wherein the processor and the memory can be connected through a bus or other manners. The memory is a non-transitory computer readable storage medium, which can be used to store a non-transitory software program, a non-transitory computer executable program and a module, and corresponding program instructions / modules. The processor executes various functions and data processing of the processor by running the non-transitory software program, instructions and modules stored in the memory, so as to implement the method for identifying subsynchronous oscillation based on VMD-MSST-HHT in the above-mentioned method embodiment.

[0057] Embodiment six, in combination Figures 3 to 7 The embodiment is used to compare the method for identifying subsynchronous oscillation with the traditional identification method, so as to fully verify the advantages of the method. Specifically, Firstly, the test signal for identifying subsynchronous oscillation used by the embodiment is as follows:

[0058] And when identifying, white noise with a signal-to-noise ratio of 20 dB needs to be added to the test signal to simulate noise interference.

[0059] Secondly, the traditional methods are short-time Fourier transform (STFT), synchronous compression transform (SST) and synchronous extraction transform (SET). Then, the time-frequency ridge effect of the above-mentioned traditional methods and the method of the embodiment for identifying subsynchronous oscillation is compared, as shown in Figure 3 As can be seen from Figure 3 (a), the limitation of the traditional STFT method is most obvious, which is specifically manifested in the following aspects: unclear ridge line: in the whole time range (0-4s), the time-frequency ridge line (representing the energy concentration area of the oscillation frequency) is in a wide strip shape and diffuses, especially the edge is unclear in the low frequency area (such as near 10-15Hz). Low resolution: it is difficult to accurately distinguish the frequency value, the low frequency energy is seriously smeared, and it is difficult to locate the exact frequency of the subsynchronous oscillation. Easy to be disturbed: there is slight noise fluctuation in the background, which further reduces the identification accuracy. From Figure 3 (b), it can be seen that the traditional SST method has improved but is still insufficient, which is specifically manifested in the following aspects: ridge line clarity is improved: compared with STFT, the ridge line is narrower and sharper, the energy is more concentrated, and the trend of frequency change with time (the ridge line is more continuous) can be tracked more clearly. But there are still obvious residual noise disturbances: there are a large number of discrete “dots” or “spur” shaped disturbances around the ridge line, these noise points will pollute the ridge line information, which will lead to misjudgment or unstable frequency tracking. The disturbance is particularly significant in the low frequency area of 5-15Hz. From Figure 3As can be seen in (c), the ridge line width is slightly wider than SST or there is diffusion in some areas, indicating that the energy concentration is not as ideal as expected. Figure 3 As can be seen in (d), the performance is the best, which is characterized by: extremely clear and sharp: the time-frequency ridge line shows the thinnest, most continuous and clearest boundary. Strong anti-noise: the background is extremely clean, ensuring the purity of the ridge line information. High frequency resolution: it can accurately lock the instantaneous frequency value of the subsynchronous oscillation (especially in the critical 15Hz area), and the resolution is much higher than the previous three. Strong robustness: in the entire time range of 0-4s, the ridge line is always stable and continuous, without breaking, blurring or jumping phenomenon, indicating that it can reliably track the oscillation process.

[0060] Finally, the above conventional methods are respectively tested by the subsynchronous oscillation test signal decomposition, like Figures 4 to 7 and the subsynchronous oscillation identification data comparison table shown in Table 1.

[0061] Table 1

[0062] As can be seen from Table 1, in terms of frequency accuracy ranking, VMD-MSST-HHT performs best: IMF2 error is only 0.006189 (the lowest of the three), and IMF3 error is 0.003240 (the second best). In terms of damping factor accuracy ranking, VMD-MSST-HH is comprehensive: IMF3 damping factor error is only 0.000353 (10 times better than the second), and IMF2 error is 0.009410 (the best), so it can be seen that the VMD-MSST-HHT method has the best comprehensive performance.

[0063] In summary, the VMD-MSST-HHT subsynchronous oscillation identification method proposed in the present application can effectively identify subsynchronous oscillation signals and solve the problems of modal aliasing, insufficient time-frequency resolution and noise sensitivity of traditional methods.

[0064] The above only describes the embodiments of the present application and does not limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the scope of the claims of the present application.

Claims

1. A subsynchronous oscillation identification method based on VMD-MSST-HHT, characterized in that: The method is: The original signal collected is decomposed into Decompose into K modal components ; For each modal component of VMD decomposition Perform multiple synchronous compression transforms (MSSTs) to obtain the MSST time-frequency distribution of each mode and then perform superposition. The superimposed result is subjected to Hilbert-Huang transform (HHT) to extract the instantaneous characteristics of the subsynchronous oscillation signal and complete the subsynchronous oscillation signal identification.

2. The subsynchronous oscillation identification method based on VMD-MSST-HHT according to claim 1, characterized in that: Original signal It is a multi-modal oscillation signal.

3. The subsynchronous oscillation identification method based on VMD-MSST-HHT according to claim 2, characterized in that: Use phasor measurement unit to collect raw signals .

4. The subsynchronous oscillation identification method based on VMD-MSST-HHT according to claim 3, characterized in that: The original signal Decompose into K modal components Specifically: The original signal Adaptive decomposition K modal components , each modal component has an instantaneous frequency ; For each modal component Perform analysis to obtain analytical signals of each modal component; Multiply the analytical signal by the estimated center frequency to shift the single-sided spectrum to the corresponding baseband; The bandwidth of each mode is estimated using Gaussian smoothing method, and a mathematical model of variational limitation of VMD is established; By performing constrained variational transformation, iterative parameter updating and optimal solution on the established mathematical model, we can obtain K The modal components IMF are used as the decomposition results of VMD.

5. The subsynchronous oscillation identification method based on VMD-MSST-HHT according to claim 1, characterized in that: Each modal component The specific steps of performing multiple synchronous compression transformation MSST are as follows: Perform synchronous compression transform SST on the original signal after VMD decomposition to obtain Short-time Fourier transform time spectrum ; right Perform a synchronous compression transformation to obtain the SST expression; The SST expression is subjected to multiple iterative synchronous compression transformations, with the number of iterations being N, to obtain the MSST time-frequency distribution of each mode.

6. The subsynchronous oscillation identification method based on VMD-MSST-HHT according to claim 1, characterized in that: The instantaneous characteristics include the instantaneous amplitude, instantaneous frequency, instantaneous phase and attenuation factor of the subsynchronous oscillation.

7. A subsynchronous oscillation identification system based on VMD-MSST-HHT, characterized in that: The system includes: Used to use variational mode decomposition VMD to collect the original signal Decompose into K modal components storage device; Each modal component used for VMD decomposition A storage device for performing multiple synchronous compression transforms (MSSTs) to obtain the MSST time-frequency distribution of each mode and then performing superposition; A storage device for performing Hilbert-Huang transform (HHT) on the superimposed result to extract the instantaneous characteristics of the subsynchronous oscillation signal and complete the identification of the subsynchronous oscillation signal.

8. The subsynchronous oscillation identification system based on VMD-MSST-HHT according to claim 7, characterized in that: The original signal Decompose into K modal components Specifically: The original signal Adaptive decomposition K modal components , each modal component has an instantaneous frequency ; For each modal component Perform analysis to obtain analytical signals of each modal component; Multiply the analytical signal by the estimated center frequency to shift the single-sided spectrum to the corresponding baseband; The bandwidth of each mode is estimated using Gaussian smoothing method, and a mathematical model of variational limitation of VMD is established; By performing constrained variational transformation, iterative parameter updating and optimal solution on the established mathematical model, we can obtain K The modal components IMF are used as the decomposition results of VMD.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the subsynchronous oscillation identification method based on VMD-MSST-HHT according to any one of claims 1 to 6 is executed.

10. A computer device, characterized in that: The device includes a memory and a processor, wherein a computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes the subsynchronous oscillation identification method based on VMD-MSST-HHT according to any one of claims 1 to 6.

Images