Fast self-adaptive damping control method for high-frequency oscillation based on frequency coupling effect

Abstract

The invention belongs to the technical field of high-voltage power transmission, and provides a high-frequency oscillation rapid self-adaptive damping control method based on a frequency coupling effect. According to the method, firstly, large errors of short time window estimation and harm possibly caused by long time delay are avoided, the response time can be shortened, meanwhile, oscillation component information can be extracted in a higher-precision and self-adaptive mode, and rapid suppression of HFO is achieved. And secondly, the influence of the frequency coupling effect on oscillation parameter estimation is fully considered, and the center frequency of the filter is adaptively modified according to the real-time estimated frequency, so that the rapid adaptive suppression of the high-frequency oscillation is realized. And finally, the high-frequency oscillation in the MMC-HVDC transmission system is quickly and effectively suppressed, the HFO can still be accurately identified even under the interference of complex conditions such as step change, and the normal operation of the system is prevented from being influenced by wrong detection.

Description

Technical Field

[0001] The present invention belongs to the technical field of high-voltage power transmission, and in particular relates to a high-frequency oscillation fast adaptive damping control method based on frequency coupling effect. Background Art

[0002] MMC-HVDC is the practical application of modular multilevel converter (MMC) in the field of high voltage direct current transmission, which has the advantages of high reliability, good waveform quality and low line loss.

[0003] However, due to the interaction between MMC and AC power grid, high-frequency oscillation (HFO) events frequently occur in the HVDC transmission system, which seriously affects the normal and stable operation of the MMC-HVDC system and even causes damage to the transmission system equipment. Therefore, studying the HFO damping control scheme is extremely important for protecting the MMC-HVDC transmission system equipment and ensuring the normal and stable operation of the power system.

[0004] The rapid suppression of high-frequency oscillations is of great significance to the normal and stable operation of the modular multilevel converter high-voltage transmission system. This process can shorten the monitoring time of HFO and avoid the oscillation events from becoming more serious.

[0005] The HFO damping control scheme of the traditional MMC-HVDC system usually connects a low-pass filter to the feedback voltage control loop to adjust the impedance characteristics of the MMC in the high-frequency band. A similar method is to reshape the impedance characteristics by optimizing the design of control parameters. However, these methods will change the response of the MMC in the entire high-frequency band, causing the protection strategy of the MMC in the high-frequency band to fail. For this problem, although a multi-tuned narrowband high-pass filter can be connected to the control loop to suppress HFO and avoid adverse effects on other irrelevant frequency bands, due to the complexity and randomness of green renewable energy generation, and the tuning frequency in the transmission system is a potential dangerous frequency that may cause the risk of oscillation instability, the operating conditions and conditions of the MMC-HVDC system may change at any time, and the oscillation frequency may also be time-varying. Therefore, the above-mentioned fixed-frequency damping control scheme is difficult to operate effectively in the case of oscillation with time-varying oscillation frequency, and an adaptive damping control scheme is particularly needed.

[0006] At present, the existing adaptive HFO damping control schemes generally connect the notch filter to the feedback voltage control loop and adaptively adjust the notch filter center frequency based on the HFO frequency estimator. These methods usually use discrete Fourier transform (DFT) to estimate the oscillation frequency, and adaptively modify the damping control parameters accordingly to ensure effectiveness under different operating conditions. However, there are usually multiple modal components in the oscillation signal, such as the dominant oscillation modal component and its coupled modal component, and the multiple frequency components of the single-phase signal spectrum are usually similar. Therefore, a longer observation time window is required in DFT to ensure the accuracy of modal extraction and frequency estimation. However, a longer time window will increase the response time of the HFO frequency estimator, so that the damping control can only be activated after the oscillation occurs for a long time, thereby affecting the normal operation of the power system. During this period of time, the oscillation amplitude of the signal may gradually increase to a larger value, thereby affecting the normal operation of the power system and may cause damage to the MMC-HVDC system equipment.

[0007] In the MMC control system, the d / q control strategy and Park transform method will introduce coupling frequency components into the HFO signal. If f0 represents the nominal frequency of the power system, f d represents the main oscillation frequency, then the Park transform and then the inverse transform will produce a coupled oscillation frequency 2f0-f d When the oscillation frequency f d When it is larger, the coupling frequency is 2f0-f d will be negative, but due to the conjugate symmetry of the DFT spectrum, |2f0-f will appear in the spectrum of the single-phase signal regardless of whether the coupling frequency is positive or negative. d |Frequency component. In HFO signal |2f0-f d | and f d The two frequencies are very close and the mutual interference is obvious, so a longer observation time window is needed to ensure the accuracy of HFO frequency estimation.

[0008] Currently, there are still few implementation schemes for fast damping control. Although these schemes have faster dynamic response, they do not consider the influence of frequency coupling. In fact, when the frequency coupling effect is significant, the results based on short time window estimation will have large errors. Summary of the invention

[0009] In view of the shortcomings of the existing phenomenon, the present invention proposes a high-frequency oscillation fast adaptive damping control method and a storage medium based on frequency coupling effect.

[0010] In one aspect, the present invention provides a method for fast adaptive damping control of high frequency oscillation based on frequency coupling effect, comprising the following steps:

[0011] Step 1: Obtain the oscillation signal of the three-phase voltage / current in the AC side of the MMC-HVDC inverter, and obtain the fundamental frequency parameters of the oscillation signal of the three-phase voltage / current.

[0012] Step 2: The three-phase voltage / current oscillation signal is processed by Clarke transform, the dynamic phasor is processed based on the frequency domain sampling theorem, and the amplitude estimation is obtained by weighted least squares processing. The oscillation frequency value preliminarily estimated by windowed interpolation DFT is combined to obtain the corrected main oscillation component frequency estimation formula, and the main oscillation component frequency and coupled oscillation component frequency estimation values ​​are obtained by solving.

[0013] Step 3: Use two band-stop filters to filter out the fundamental component and coupled oscillation component of the signal respectively, extract the main oscillation component signal through a bandpass filter, shift the phase to obtain the reference current signal, adjust the amplitude according to the presence or absence of HFO, obtain the limited reference current signal through a limiter, and then generate a damping current through a voltage source converter.

[0014] Step 4: Inject the damping current into the common coupling point PCC on the AC side of the MMC-HVDC inverter to suppress HFO.

[0015] On the other hand, the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and the computer program is used to execute the above-mentioned high-frequency oscillation fast adaptive damping control method based on frequency coupling effect.

[0016] Beneficial effects of the present invention:

[0017] (1) The present invention avoids the large error of short time window estimation and the harm that may be caused by long time delay, and can extract the oscillation component information with higher accuracy and adaptiveness while shortening the response time, so as to achieve rapid suppression of HFO.

[0018] (2) The present invention fully considers the influence of frequency coupling effect on oscillation parameter estimation, and adaptively modifies the filter center frequency according to the real-time estimated frequency to achieve rapid adaptive suppression of high-frequency oscillations.

[0019] (3) The present invention can quickly and effectively suppress high-frequency oscillations in the MMC-HVDC transmission system. Even under the interference of complex conditions such as step changes, HFO can still be accurately identified, thus avoiding the impact of erroneous detection on the normal operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of a method for fast adaptive damping control of high-frequency oscillations based on frequency coupling effect according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The present invention provides a method for fast adaptive damping control of high-frequency oscillation based on frequency coupling effect, comprising the following steps:

[0023] Step 1: Obtain the oscillation signal of the three-phase voltage / current in the AC side of the MMC-HVDC inverter, and obtain the fundamental frequency parameters of the oscillation signal of the three-phase voltage / current.

[0024] Step 2: The three-phase voltage / current oscillation signal is processed by Clarke transform, the dynamic phasor is processed based on the frequency domain sampling theorem, and the amplitude estimation is obtained by weighted least squares processing. The oscillation frequency value preliminarily estimated by windowed interpolation DFT is combined to obtain the corrected main oscillation component frequency estimation formula, and the main oscillation component frequency and coupled oscillation component frequency estimation values ​​are obtained by solving.

[0025] Step 3: Use two band-stop filters to filter out the fundamental component and coupled oscillation component of the signal respectively, extract the main oscillation component signal through a bandpass filter, shift the phase to obtain the reference current signal, adjust the amplitude according to the presence or absence of HFO, obtain the limited reference current signal through a limiter, and then generate a damping current through a voltage source converter.

[0026] Step 4: Inject the damping current into the common coupling point PCC on the AC side of the MMC-HVDC inverter to suppress HFO.

[0027] Further, the step 1 specifically includes:

[0028] Step 1-1: Obtain the oscillation signal of the three-phase voltage / current in the AC side of the MMC-HVDC inverter.

[0029] Step 1-2: Abstractly express the oscillation signal of the three-phase voltage / current as formula (1):

[0030]

[0031] Among them, when x represents the A phase signal, x=0 in 2πx / 3; when x represents the B phase, x=-1; when x represents the C phase, x=1;

[0032] A q,x represents the initial amplitude of the q component of the signal in the x phase, f q,xrepresents the initial phase of the q component of the signal in the x phase (the subscript q is 0 for the fundamental frequency component, d for the main oscillation component, and p for the coupled oscillation component), and σ is the damping factor of the HFO.

[0033] Step 1-3: Obtain the fundamental frequency of the oscillation signal of the three-phase voltage / current.

[0034] Further, the step 2 specifically includes:

[0035] Step 2-1: The three-phase voltage / current oscillation signal is subjected to Clarke transformation to obtain a phasor expression of the three-phase voltage / current oscillation signal.

[0036] Step 2-2: Approximate the phasor expression as the sum of several imaginary exponential functions to obtain a matrix expression of the three-phase voltage / current oscillation signal.

[0037] Step 2-3: Perform weighted least squares processing on the matrix expression to obtain a weighted least squares estimation formula for the phasor samples in the oscillation signal matrix expression.

[0038] Step 2-4: Combine the weighted least squares estimation formula with the derivative estimation formula of the imaginary exponential function and obtain an amplitude estimate, and then combine it with the oscillation frequency value initially estimated by the windowed interpolation DFT to obtain a corrected main oscillation component frequency estimation formula, and solve to obtain the main oscillation component frequency and coupled oscillation component frequency estimation values.

[0039] Further, the step 3 specifically includes:

[0040] Step 3-1: The fundamental frequency component and the coupled oscillation component are successively filtered out by a band-stop filter, and then extracted by a band-pass filter to obtain a main oscillation component signal.

[0041] Step 3-2: Shift the phase of the main oscillation component signal through a phase shifter to obtain a reference current signal.

[0042] Step 3-3: adjusting the amplitude of the reference current signal through a gain controller according to whether there is an HFO event, to obtain an amplitude-adjusted reference current signal.

[0043] Step 3-4: The amplitude-adjusted reference current signal is limited to a maximum and a minimum amplitude by a limiter to obtain a limited reference current signal.

[0044] Step 3-5: Inject the limited reference current signal into the voltage source converter to obtain a damping current.

[0045] Furthermore, in step 2-1, the Clarke transformation of equation (1) is performed to obtain:

[0046]

[0047] Among them, the subscript α indicates that the corresponding variable is in positive order;

[0048] After the three-phase signal is transformed by Clarke, only the fundamental component, the main oscillation component and the coupled oscillation component exist in the spectrum.

[0049] Furthermore, the S-domain transfer functions of the band-stop filter and the band-pass filter in step 3-1 are respectively:

[0050]

[0051] Wherein, formula (3) represents two second-order band-stop filters at the same time, and formula (4) represents a second-order band-pass filter; s is the complex frequency, which contains the angular frequency information; f c represents the stopband center frequency, which is the fundamental component frequency f0 and the coupled oscillation component frequency 2f0-f in the two band-stop filters. d ; ξ is the damping ratio; f d is the oscillation frequency.

[0052] Furthermore, the phase shifter in step 3-2 is specifically a second-order phase shifter obtained by cascading two first-order improved phase shifters with unit gain.

[0053] Furthermore, the HFO event determination is specifically as follows: when the amplitude estimation value of the main oscillation component is greater than a threshold value, and its first-order derivative is within a range from 0 to a set amplitude change rate threshold value, the HFO damping control is activated.

[0054] Furthermore, in step 3-3, the amplitude adjustment is performed by the gain controller according to whether or not there is HFO:

[0055] When the HFO damping control is activated, the gain of the reference current signal is adjusted to a preset value;

[0056] When there is no HFO event, the gain is set to 0.

[0057] The implementation steps of the present invention are further described below with reference to the accompanying drawings and specific examples:

[0058] like Figure 1 As shown, the embodiment of the present application provides a method for fast adaptive damping control of high-frequency oscillation based on frequency coupling effect, which specifically includes the following steps:

[0059] Step 1: Obtain the oscillation signal of the three-phase voltage / current in the AC side of the MMC-HVDC inverter, and obtain the fundamental frequency parameters of the oscillation signal of the three-phase voltage / current.

[0060] Preferably, step 1 specifically includes:

[0061] Step 1-1: Obtain the oscillation signal of the three-phase voltage / current in the AC side of the MMC-HVDC inverter;

[0062] Step 1-2: Abstractly express the oscillation signal of the three-phase voltage / current as formula (1):

[0063]

[0064] When x represents the A phase signal, x = 0 in 2πx / 3; when x represents the B phase, x = -1; when x represents the C phase, x = 1; A q,x represents the initial amplitude of the q component of the signal in the x phase (the subscript q is 0 for the fundamental frequency component, d for the main oscillation component, and p for the coupled oscillation component); f q,x represents the initial phase of the q component of the signal in the x phase; σ is the damping factor of HFO.

[0065] Step 1-3: Obtain the fundamental frequency f0 of the oscillation signal of the three-phase voltage / current.

[0066] Step 2: The three-phase voltage / current oscillation signal is processed by Clarke transform, the dynamic phasor is processed based on the frequency domain sampling theorem, and the amplitude estimation is obtained by weighted least squares processing. The oscillation frequency value preliminarily estimated by windowed interpolation DFT is combined to obtain the corrected main oscillation component frequency estimation formula, and the main oscillation component frequency and coupled oscillation component frequency estimation values ​​are obtained by solving.

[0067] Preferably, step 2 specifically includes:

[0068] Step 2-1: Process the three-phase voltage / current oscillation signal through Clarke transformation to obtain the phasor expression of the three-phase voltage / current oscillation signal.

[0069] Since the integrated signal after Clarke transformation is a complex function, its DFT spectrum does not have conjugate symmetry and can show negative frequency components in the spectrum. Compared with direct DFT, the spectrum is sparser and less affected by the coupling frequency, so the main oscillation frequency component can be estimated more quickly.

[0070] Since the power system is usually three-phase balanced, that is, the amplitudes of the corresponding frequency components are equal, the frequencies are equal, and the phases differ by 2π / 3. Clarke transformation of equation (1) yields:

[0071]

[0072] The subscript α indicates that the corresponding variable is in positive order.

[0073] It can be clearly seen from formula (2) that after the three-phase signal is transformed by Clarke, only the fundamental component, the main oscillation component and the coupled oscillation component exist in the spectrum. Compared with the original signal, the processed signal will not have conjugate symmetric components in the spectrum, that is, each frequency component can only be reflected at the spectrum position corresponding to them. Therefore, the frequency spacing of each component will become larger after processing, which effectively reduces the mutual interference between frequencies and can estimate the oscillation parameters more quickly.

[0074] Step 2-2: Approximately express the phasor expression of the three-phase voltage / current oscillation signal as the sum of several imaginary exponential functions based on the frequency domain sampling theorem to obtain a matrix expression of the three-phase voltage / current oscillation signal.

[0075] Define the fundamental frequency phasor as The main oscillating phase is The coupled oscillating phasor is Formula (2) can be further expressed as formula (3):

[0076]

[0077] Due to the existence of the damping factor σ, the amplitudes of the oscillating component and its coupled component exhibit dynamic time-varying characteristics.

[0078] Based on the frequency domain sampling theorem, the dynamic phasor can be approximately represented by several imaginary functions within a limited observation window, as shown in formula (4):

[0079]

[0080] Where k∈Z, t∈[-T, T], 2T and 2K represent the window length and order respectively; Δf is the frequency domain sampling interval; G q,k are samples in the frequency domain (q∈{0,d,p}).

[0081] Step 2-3: Perform weighted least squares processing on the matrix expression of the three-phase voltage / current oscillation signal to obtain a weighted least squares estimation formula of the phasor sample G in the matrix expression of the oscillation signal.

[0082] Using sampling frequency f s After time domain sampling, there are 2N+1 sample points in the time window n∈[-N,N]. Formula (3) can be expressed in matrix form as Formula (5):

[0083]

[0084] In the formula, S is the α,β (t) is a column vector of 2N+1 samples; E is a column vector containing The matrix of sample information, where q∈{0,d,p}, f p=2f0-f d ; G is a vector sample G q,k Column vector of .

[0085] The phasor sample G in formula (5) can be estimated using the weighted least squares method (WLS), and the estimation formula is shown in formula (6):

[0086]

[0087] Wherein, the superscript H represents the Hermitian operator; W is the weight matrix containing the window function information.

[0088] Step 2-4: Combine the weighted least squares estimate of the phasor sample G in the oscillation signal matrix expression with the derivative estimate of the imaginary exponential function and to obtain the amplitude estimate |g d (t)|, and then combined with the oscillation frequency value preliminarily estimated by windowed interpolation DFT, a modified main oscillation component frequency estimation formula is obtained. The modified main oscillation component frequency estimation formula is solved to obtain the main oscillation component frequency and the coupled oscillation component frequency estimation values.

[0089] From formula (4), we can see that the phasor estimation formulas of the main oscillation component of the signal and its first-order derivative at time t = 0 are:

[0090]

[0091] Based on the frequency measurement estimation formula defined in the IEEE / IEC 60255-118-1 standard, the frequency of HFO can be estimated by equations (7) and (8), as follows:

[0092]

[0093] Among them, Im[·] means taking the imaginary part of the corresponding part; * means taking the conjugate of the corresponding part.

[0094] Formula (9) can be regarded as the existing oscillation frequency f d The actual frequency correction of f d is unknown, so a preliminary estimate is needed before accurately estimating the frequency.

[0095] From formula (2), we can see that S α,β The mutual interference between the frequency components is small, so the embodiment of the present application can perform a three-line interpolation FFT with a Hanning window based on a short observation time window to preliminarily estimate the oscillation frequency f d After that, the coupled oscillation frequency can also be calculated by the frequency estimate obtained by equation (9), that is,

[0096] If we perform a modulo operation on equation (4), we can know that the oscillation amplitude |g d (t)| is actually the result of formula (7) The calculation required It can be obtained by formula (6), thus realizing the estimation of HFO amplitude.

[0097] Step 3: Use two band-stop filters to filter out the fundamental component and coupled oscillation component of the signal respectively, extract the main oscillation component signal through a bandpass filter, shift the phase to obtain the reference current signal, adjust the amplitude according to the presence or absence of HFO, obtain the limited reference current signal through a limiter, and the voltage source converter generates a damping current.

[0098] Preferably, step 3 specifically includes:

[0099] Step 3-1: The fundamental frequency component and the coupled oscillation component are successively filtered out by a band-stop filter, and then extracted by a band-pass filter to obtain a main oscillation component signal.

[0100] In order to obtain the main oscillation component of the signal and its characteristics, it is necessary to first use two band-stop filters to filter out the fundamental component and the coupled oscillation component of the signal respectively, and then extract the main oscillation component through a band-pass filter. The S-domain transfer functions of the band-stop filter and the band-pass filter designed in the embodiment of the present application are:

[0101]

[0102] Formula (10) represents two second-order band-stop filters at the same time, and formula (11) represents a second-order band-pass filter. Where s is the complex frequency, which contains the angular frequency information; f c represents the stopband center frequency, which is the fundamental component frequency f0 and the coupled oscillation component frequency 2f0-f in the two band-stop filters. d ; ξ is the damping ratio; f d is the oscillation frequency.

[0103] Since the stopband center frequency f of the above filter c and the oscillation frequency f d It will change with the estimated frequency, so it can realize adaptive tracking filtering of the filter, which not only ensures the adequacy of filtering out other components, but also ensures the accuracy of extracting the target component, greatly reducing the error caused by frequency offset.

[0104] Step 3-2: Shift the phase of the main oscillation component signal through a phase shifter to obtain a reference current signal.

[0105] Assume that the main oscillation signal extracted by the bandpass filter is u d , through the phase shifter u d The phase of the reference current signal i is shifted and generatedd The embodiment of the present application uses a second-order phase shifter obtained by cascading two first-order improved phase shifters with unit gain, and its transfer function is:

[0106]

[0107] Where, T c is a time constant that can be changed by changing T c The size of the signal i is changed d In order to ensure the effectiveness of HFO suppression, it is necessary to make i d with u abc The phase difference of the original dominant oscillating signal in is close to or less than π / 2. Note that a certain safety margin needs to be reserved here to avoid the oscillation aggravation caused by the introduction of a phase shift greater than π / 2 in some cases. In short, this phase difference is equal to the phase delay in two band-stop filters, a band-pass filter and the phase shifter, and then Brent's method can be used to calculate T c .

[0108] Step 3-3: adjusting the amplitude of the reference current signal through a gain controller according to whether there is an HFO event, to obtain an amplitude-adjusted reference current signal.

[0109] Specifically, when the HFO damping control is activated, the reference current signal i d The gain will be adjusted to a preset value c; when there is no HFO event, the gain is set to 0 to avoid unnecessary control.

[0110] To determine whether an HFO event occurs, the oscillation detection threshold A can be set. th , and compare it with the amplitude estimate of the main oscillating component in the signal |g d (t)| is done by comparison.

[0111] However, in actual situations, there may be an overshoot phenomenon where the amplitude suddenly increases and quickly exceeds the threshold, such as a step change in the signal. To avoid false detection in similar situations, the amplitude change rate threshold D can be set. th , and compare it with the first-order derivative of the amplitude estimate. Specifically, the first-order derivative of the amplitude estimate is obtained by the ratio of the difference between the two amplitude estimates before and after the detection time to the time difference between them, which is greater than zero and less than D th It meets the characteristics of HFO events.

[0112] In summary, the HFO event determination is specifically as follows: when the amplitude estimate of the main oscillation component is greater than the threshold A th , and its first-order derivative is between 0 and D thHFO damping control is activated within this range.

[0113] Step 3-4: The amplitude-adjusted reference current signal is limited to a maximum and a minimum amplitude by a limiter to obtain a limited reference current signal.

[0114] Use the amplitude control function of the limiter to limit i d By measuring the maximum and minimum amplitudes of the

[0115] Step 3-5: Inject the limited reference current signal into the voltage source converter (VSC) to generate the HFO damping current i ADC The purpose of suppressing HFO is achieved by injecting damping current into the MMC-HVDC transmission system through the common coupling point (PCC) on the AC side of the inverter.

[0116] Step 4: Inject the damping current into the common coupling point PCC on the AC side of the MMC-HVDC inverter to suppress HFO.

[0117] The embodiment of the present application also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a high-frequency oscillation fast adaptive damping control method based on frequency coupling effect are implemented. Specifically, the computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include a read-only memory (ROM), a hard disk, a flash memory, an optical disk, a magnetic disk, etc.

[0118] Finally, it should be noted that the purpose of publishing the embodiments is to help further understand the present invention, but those skilled in the art can understand that various substitutions and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the present invention should not be limited to the contents disclosed in the embodiments, and the scope of protection claimed by the present invention shall be subject to the scope defined in the claims.

Claims

1. A method for fast adaptive damping control of high-frequency oscillation based on frequency coupling effect, characterized in that: The following steps are involved: Step 1: Obtain the oscillation signal of the three-phase voltage / current in the AC side of the MMC-HVDC inverter, and obtain the fundamental frequency parameters of the oscillation signal of the three-phase voltage / current; Step 2: The three-phase voltage / current oscillation signal is processed by Clarke transformation, the dynamic phasor is processed based on the frequency domain sampling theorem, and the amplitude estimation is obtained by weighted least squares processing. The oscillation frequency value preliminarily estimated by windowed interpolation DFT is combined to obtain the corrected main oscillation component frequency estimation formula, and the main oscillation component frequency and coupled oscillation component frequency estimation values ​​are obtained by solving; Step 3: Use two band-stop filters to filter out the fundamental component and coupled oscillation component of the signal respectively, extract the main oscillation component signal through a band-pass filter, shift the phase to obtain a reference current signal, adjust the amplitude according to whether there is HFO, obtain the limited reference current signal through a limiter, and then generate a damping current through a voltage source converter; Step 4: Inject the damping current into the common coupling point PCC on the AC side of the MMC-HVDC inverter to suppress HFO.

2. The method for fast adaptive damping control of high-frequency oscillation based on frequency coupling effect according to claim 1, characterized in that: The step 1 specifically includes: Step 1-1: Obtain the oscillation signal of the three-phase voltage / current in the AC side of the MMC-HVDC inverter; Step 1-2: Abstractly express the oscillation signal of the three-phase voltage / current as formula (1): Among them, when x represents the A phase signal, x=0 in 2πx / 3; when x represents the B phase, x=-1; when x represents the C phase, x=1; A q,x represents the initial amplitude of the q component of the signal in the x phase, f q,x represents the initial phase of the q component of the signal in the x phase, the subscript q is 0 for the fundamental frequency component, d for the main oscillation component, p for the coupled oscillation component, and σ is the damping factor of the HFO; Step 1-3: Obtain the fundamental frequency of the oscillation signal of the three-phase voltage / current.

3. The method for fast adaptive damping control of high-frequency oscillation based on frequency coupling effect according to claim 1, characterized in that: The step 2 specifically includes: Step 2-1: Perform Clarke transformation on the three-phase voltage / current oscillation signal to obtain a phasor expression of the three-phase voltage / current oscillation signal; Step 2-2: Approximately express the phasor expression as the sum of several imaginary exponential functions to obtain a matrix expression of the three-phase voltage / current oscillation signal; Step 2-3: Perform weighted least squares processing on the matrix expression to obtain a weighted least squares estimation formula of the phasor samples in the oscillation signal matrix expression; Step 2-4: Combine the weighted least squares estimation formula with the derivative estimation formula of the imaginary exponential function and obtain an amplitude estimate, and then combine it with the oscillation frequency value initially estimated by the windowed interpolation DFT to obtain a corrected main oscillation component frequency estimation formula, and solve to obtain the main oscillation component frequency and coupled oscillation component frequency estimation values.

4. The method for fast adaptive damping control of high frequency oscillation based on frequency coupling effect according to claim 1, characterized in that: The step 3 specifically includes: Step 3-1: The fundamental frequency component and the coupled oscillation component are successively filtered out by a band-stop filter, and then extracted by a band-pass filter to obtain a main oscillation component signal; Step 3-2: Shifting the main oscillation component signal by a phase shifter to obtain a reference current signal; Step 3-3: adjusting the amplitude of the reference current signal through a gain controller according to whether there is an HFO event, to obtain an amplitude-adjusted reference current signal; Step 3-4: limiting the maximum and minimum amplitudes of the amplitude-adjusted reference current signal through a limiter to obtain a limited reference current signal; Step 3-5: Inject the limited reference current signal into the voltage source converter to obtain a damping current.

5. The method for fast adaptive damping control of high frequency oscillation based on frequency coupling effect according to claim 2, characterized in that: In step 2-1, the Clarke transformation of equation (1) is performed to obtain: Among them, the subscript α indicates that the corresponding variable is in positive order; After the three-phase signal is transformed by Clarke, only the fundamental component, the main oscillation component and the coupling frequency component exist in the spectrum.

6. The method for fast adaptive damping control of high frequency oscillation based on frequency coupling effect according to claim 4, characterized in that: The S-domain transfer functions of the band-stop filter and the band-pass filter in step 3-1 are: Wherein, formula (3) represents two second-order band-stop filters at the same time, and formula (4) represents a second-order band-pass filter; s is the complex frequency, which contains the angular frequency information; f c represents the stopband center frequency, which is the fundamental component frequency f0 and the coupled oscillation component frequency 2f0-f in the two band-stop filters. d ; ξ is the damping ratio; f d is the oscillation frequency.

7. The method for fast adaptive damping control of high frequency oscillation based on frequency coupling effect according to claim 4, characterized in that: The phase shifter in step 3-2 is specifically a second-order phase shifter obtained by cascading two first-order improved phase shifters with unit gain.

8. The method for fast adaptive damping control of high frequency oscillation based on frequency coupling effect according to claim 4, characterized in that: The HFO event determination is specifically as follows: When the amplitude estimate of the main oscillation component is greater than a threshold value and its first-order derivative is within the interval from 0 to a set amplitude change rate threshold value, the HFO damping control is activated.

9. The method for fast adaptive damping control of high frequency oscillation based on frequency coupling effect according to claim 8, characterized in that: In step 3-3, the amplitude is adjusted by the gain controller according to whether or not there is HFO: When the HFO damping control is activated, the gain of the reference current signal is adjusted to a preset value; When there is no HFO event, the gain is set to 0.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the high-frequency oscillation fast adaptive damping control method based on frequency coupling effect as described in any one of claims 1 to 9 above.

Images