A method and device for suppressing high-frequency oscillation of a doubly-fed fan by stator voltage feedforward

Abstract

The application discloses a double-fed fan high-frequency oscillation suppression method and device of stator voltage feedforward. The method comprises the following steps: introducing a stator voltage feedforward channel into the control loop of a grid-side converter and a rotor-side converter at the same time, setting a feedforward point before a digital control delay link, and configuring a unified virtual impedance coefficient for the two sides; and remodeling the overall output impedance of the system as Zs through the above structure. The application makes the impedance remodeling equivalent coefficient only related to the virtual impedance and the digital control delay, completely decoupled from the fan body parameters, and a same set of suppressor can be adapted to different models of units without redesigning, so that adaptive suppression without frequency detection is realized. In combination with phase compensation and filtering, the application can effectively suppress high-frequency oscillation caused by parallel capacitor switching or unit switching, and significantly improves the wideband stability and engineering applicability of the grid-connected system.

Description

Technical Field

[0001] This invention relates to the field of power system stability and control technology, specifically to a method and apparatus for suppressing high-frequency oscillations in a doubly fed wind turbine grid-connected system under weak grid conditions, particularly applicable to AC grid scenarios including parallel compensation capacitors. Background Technology

[0002] With the large-scale grid connection of wind power and the increase in grid transmission distance, the power grid exhibits weak grid characteristics, leading to increasingly prominent high-frequency oscillation problems that seriously threaten the safe and stable operation of wind farms and the power grid. Existing high-frequency oscillation suppression methods are mainly divided into two categories: passive damping and active damping. Among active damping methods, control strategies based on virtual impedance have become a research hotspot due to their flexibility and lack of the need for additional hardware.

[0003] However, existing technologies still have the following significant shortcomings:

[0004] (1) Some suppression methods based on stator voltage feedforward usually only control one side (such as the rotor side) converter and do not fully consider the cooperative suppression requirements of the grid-side converter and the rotor-side converter at high frequencies. Since the grid-side converter and the rotor-side converter of the doubly fed wind turbine are coupled through the DC bus, their high-frequency impedance characteristics affect each other, and it is difficult to achieve effective reshaping of the overall output impedance by single-side control.

[0005] (2) The design of existing virtual impedance methods heavily relies on precise matching at specific resonant frequencies. When the value of the parallel compensation capacitor on the grid side changes significantly (e.g., from 3μF to 11μF), the system resonant frequency drifts accordingly, and the original suppression strategy is prone to failure. This necessitates the configuration of complex frequency detection and tracking circuits in the suppressor, increasing system cost and implementation difficulty.

[0006] (3) The control performance of existing suppressors is highly sensitive to the parameters of the unit itself (such as rotor resistance, inductance, etc.). The parameters of doubly fed wind turbines of different models and capacities vary significantly, which means that the same set of suppression parameters cannot be used universally. The proportional-integral coefficient of the controller must be redesigned or even readjusted for each unit, resulting in a huge workload for engineering adaptation.

[0007] (4) In the relevant technical fields, some high-frequency oscillation suppression schemes for flexible DC transmission grid connection or inverter grid connection have fundamentally different control mechanisms and parameter design principles from the weak AC grid scenario with parallel compensation targeted by this patent, and are difficult to apply directly.

[0008] Therefore, there is an urgent need in this field for a high-frequency oscillation suppression method that can get rid of the dependence on real-time resonant frequency detection, has wide bandwidth adaptability, and does not require redesigning suppression parameters due to differences in wind turbine models or parameters. Summary of the Invention

[0009] The present invention aims to overcome the above-mentioned shortcomings of the prior art and provide a method and device for suppressing high-frequency oscillations of doubly fed wind turbines that is highly adaptive, requires no frequency detection, and is independent of specific unit parameters, so as to solve the technical problems of strong parameter dependence and weak adaptability to changes in operating conditions in the prior art.

[0010] To achieve the above objectives, the present invention adopts the following technical solution:

[0011] In a first aspect, the present invention provides a method for suppressing high-frequency oscillations in a doubly-fed wind turbine with stator voltage feedforward, characterized by comprising the following steps:

[0012] Step S1: Simultaneously introduce stator voltage feedforward channels into the control loops of the grid-side converter (GSC) and the rotor-side converter (RSC).

[0013] Step S2, set the feedforward point of the stator voltage feedforward channel in the digital control delay stage ( )Before;

[0014] Step S3: In the stator voltage feedforward channel, configure a unified virtual impedance coefficient for the grid-side converter and the rotor-side converter. That is, satisfying ,in For the virtual impedance on the GSC side, This is the virtual impedance on the RSC side;

[0015] Step S4, through the stator voltage feedforward and the unified virtual impedance coefficient The overall output impedance of the doubly fed wind turbine grid-connected system is reshaped to The form;

[0016] in, For impedance reshaping equivalent coefficient, The output impedance before reshaping. This is a digital control delay transfer function.

[0017] Optionally, the virtual impedance coefficient It can be implemented in various forms, such as proportional coefficient, first-order high-pass filter, first-order low-pass filter or proportional-differential element, to adapt to different system delay characteristics and oscillation frequency bands.

[0018] Furthermore, prior to step S1, the method further includes an optional preparatory step S0: establishing an impedance model of the doubly-fed wind turbine grid-connected system based on impedance analysis; and quantifying the control delay using variable-weight grey relational analysis. , and add capacitor By analyzing the sensitivity of multiple influencing factors, including those affecting the system's resonant frequency, the strongly correlated factors that play a dominant role in high-frequency oscillations were identified. This step provides a theoretical basis for the design of the control structure.

[0019] Furthermore, the method also includes step S5: connecting an oscillation suppressor in series in the stator voltage feedforward channel. The oscillation suppressor is composed of a phase compensator and a second-order Chebyshev filter cascaded together to further compensate for phase, filter out noise, and enhance system robustness.

[0020] The working principle and core innovation of this invention are as follows:

[0021] This invention creatively combines three features: "simultaneous stator voltage feedforward introduction on both sides," "shifting the feedforward point before the delay stage," and "using a unified virtual impedance coefficient on both sides." Through this specific control structure, and after rigorous impedance modeling and block diagram equivalent transformation, the reshaping expression for the overall output impedance of the doubly-fed induction generator (DFIG) is derived as follows: ,in .

[0022] The key point is: The expression is only related to the virtual impedance selected by the designer. and the inherent digital control delay of the system Related to, but not including at all, the rotor resistance of the doubly fed wind turbine body. Rotor inductance Stator resistance Stator inductor and mutual induction Electromechanical parameters. This means the parameters of the suppressor designed for a specific type of unit. It can be applied directly to another type of unit without any modification, fundamentally solving the parameter dependency problem of existing methods and realizing a "plug-and-play" universal design.

[0023] Because the impedance reshaping behavior of this invention does not depend on matching to a fixed resonant frequency, but rather on... The coefficient scales the output impedance consistently across the entire control frequency band, therefore, when the grid is connected in parallel with a compensation capacitor... Even when significant changes cause the system's resonant frequency to drift, the method of this invention can still maintain sufficient phase margin (measured ≥34°) over a wide frequency range. This eliminates the need for complex frequency detection and tracking mechanisms, achieving adaptive wideband oscillation suppression.

[0024] This method has been verified in practical engineering. When high-frequency oscillations occurred in a 220kV power grid in western province, the application of this method rapidly reduced the total harmonic distortion (THD) at the point of common coupling from 56.13% to 1.35% within 60 milliseconds, and the system quickly returned to stability. Simulations and experiments both show that the same set of parameters can achieve excellent suppression effects (THD < 1%) on both 2MW and 7.5kW generating units of different capacities.

[0025] The entire solution is based on an improved control algorithm and can be fully embedded in the digital controller (such as DSP or FPGA) of the existing converter. It requires no additional hardware and has extremely low modification costs, making it easy to promote in existing wind farms.

[0026] Secondly, the present invention provides an apparatus for implementing the above-described method, characterized in that it includes a controller reconfiguration module and an impedance reshaping module, and may further include a grey relational analysis module, a filtering and compensation module, a data acquisition unit, etc. This apparatus can be integrated into the digital control unit of a wind turbine converter.

[0027] Thirdly, the present invention provides a doubly fed wind turbine converter control system including the above-described device. Attached Figure Description

[0028] Figure 1 This is a flowchart of the method of the present invention.

[0029] Figures 2a-2c The block diagram of the control structure transformation for the parallel virtual impedance on the GSC side is shown below. Figure 2a For the original control structure, Figure 2b To introduce stator voltage feedforward and shift the feedforward point backward, Figure 2c This is a block diagram of impedance reshaping after equivalent transformation.

[0030] Figures 3a-3c The control structure transformation block diagram for the parallel virtual impedance on the RSC side is shown, and the feedforward introduction and equivalent transformation process are similarly illustrated.

[0031] Figure 4 This is a schematic diagram illustrating the parameter constraints design of the filter in the oscillation suppressor.

[0032] Figures 5a-5b The impedance Bode plots are obtained by reshaping two different capacity units, 2MW and 7.5kW, using the method of this invention.

[0033] Figures 6a-6b The graph shows a comparison of the impedance reshaping effect of the traditional virtual impedance method on the two types of units mentioned above, demonstrating that the traditional method fails after unit switching.

[0034] Figure 7The impedance reshaping curves of the method of the present invention under different parallel compensation capacitors (3μF to 11μF) show the wideband adaptability.

[0035] Figure 8 This invention provides a comparison of the phase margin before and after impedance reshaping of a 2MW unit using the method of this invention.

[0036] Figures 9a-9b This invention provides a comparison of the phase margin before and after impedance reshaping of a 7.5kW unit using the method of this invention.

[0037] Figure 10 This is a schematic diagram of the topology of a 220kV power system in the western part of a province and the location of oscillation monitoring points.

[0038] Figures 11a-11b This invention provides a comparison of the actual power system operating waveforms before and after the implementation of the method. Figure 11a The waveform before input is distorted. Figure 11b The waveform is a sine wave after 60ms input. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below.

[0040] Example 1

[0041] This embodiment details a method for suppressing high-frequency oscillations in a doubly fed wind turbine with stator voltage feedforward, provided by the present invention.

[0042] Step S0: System Analysis and Identification of Dominant Factors (Recommended Preparatory Step)

[0043] First, a Norton equivalent model of a doubly-fed induction generator (DFIG) grid-connected system is established based on impedance analysis. The DFIG is equivalent to a controlled current source and its output impedance. In parallel connection, the power grid is equivalent to an ideal voltage source and the grid impedance. Series connection. The grid impedance includes the grid-side equivalent resistance. Equivalent inductance and parallel compensation capacitors : The output impedance characteristics of the system are obtained by frequency sweeping by injecting a small perturbation voltage into the common coupling point and measuring the response current.

[0044] Secondly, variable-weight grey relational analysis is used to quantify the effect of each factor on the resonant frequency. The impact. Selection includes control delay. , and add capacitor , grid-side inductor Grid-side resistance In addition, several characteristic factors such as the converter current inner loop PI parameters are considered. An analysis matrix is ​​constructed, and the first... The factor in the first The value under this working condition is The resonant frequency is .

[0045] Define sensitivity metrics:

[0046]

[0047] in , These are the resonant frequency and factor values ​​under the reference operating conditions, respectively. This refers to the number of operating conditions.

[0048] Normalize the original data and calculate the correlation coefficient. Then, using sensitivity indicators Assuming the weights of each factor, calculate the weighted grey relational degree. .

[0049] The calculation results clearly show: control delay The correlation is approximately 0.2509, and the capacitor is supplemented. The correlation coefficient between the two factors was approximately 0.5157, significantly higher than that of other factors (all below 0.1), thus identifying them as strongly correlated factors causing high-frequency oscillations. This analysis provides a clear direction for subsequent targeted compensation designs for control delay.

[0050] Steps S1-S4: Core control structure reconstruction and impedance reshaping

[0051] Based on the above analysis, the controllable factor among the strongly correlated factors is the controllable delay. Innovate control structures.

[0052] like Figure 2a As shown, the original GSC control structure includes a current inner loop, voltage feedforward, and modulation delay. This invention introduces a stator voltage into the GSC control loop. The feedforward channel. The key innovation lies in: such as Figure 2b As shown, the injection point of the feedforward signal is moved from the traditional delay stage. Then, the process is moved to the delayed phase. Previously, this "feedforward shift" design enabled subsequent virtual impedance to compensate for the entire control path, including delays.

[0053] Inserting virtual impedance in the feedforward path One of the core settings of this invention is to make the virtual impedance on the GSC side... Virtual impedance on the RSC side Equal, that is, using a uniform virtual impedance coefficient. This symmetrical design ensures the consistency of impedance reshaping behavior on both sides of the converter, which is key to achieving a concise expression of the overall impedance.

[0054] After strict equivalent transformation of the block diagram (such as...) Figure 2c , Figures 3a-3c As shown in the figure, the reshaped GSC-side output impedance can be obtained. The reshaped RSC-side output impedance Since the GSC and RSC are connected in parallel at the common coupling point, the output impedance of the entire doubly-fed induction generator (DFIG) is:

[0055]

[0056] in:

[0057]

[0058] For digital control delay transfer functions, they can typically be modeled as follows: Alternatively, Pade can be approximated as a first-order inertial element. , This is the delay time (generally 1.5 to 2 switching cycles).

[0059] from The expression can be seen intuitively: The value depends only on the designer's choice. and the inherent system This is completely independent of the rotor resistance, inductance, and other intrinsic parameters of the doubly-fed wind turbine. This mathematically proves the "parameter decoupling" characteristic of this invention.

[0060] Step S5: Oscillation Suppressor Detailed Design (Optimization Step)

[0061] To further improve the suppression effect and system robustness, an oscillation suppressor is connected in series in the feedforward channel. .

[0062] Phase compensator Design:

[0063] Using a second-order look-ahead network ( Its transfer function is:

[0064]

[0065] In this embodiment, the following is selected: , This selection provides a maximum phase lead of approximately 60° in the 500Hz to 1500Hz frequency band to compensate for phase lag introduced by control delay and filters. It should be noted that... and It can be adjusted within a certain range according to the oscillation frequency band of the actual system: generally , and maintain To ensure a sufficient lead angle.

[0066] The digital implementation of the phase compensator can employ a bilinear transform: Let ,in For the sampling period, substitute... The z-domain transfer function is then obtained and converted into a difference equation.

[0067] Second-order Chebyshev filter Design:

[0068] The function of a filter is to provide a fast roll-off outside the effective suppression frequency band to suppress high-frequency switching noise and measurement noise. This embodiment uses a second-order Chebyshev low-pass filter with the following design parameters:

[0069] Passband ripple factor (Corresponding to a passband ripple of approximately 0.05 dB);

[0070] Cutoff frequency This frequency is typically set to 1.2 to 1.5 times the upper limit of the target oscillation frequency band;

[0071] According to the Chebyshev filter design formula, the pole locations are:

[0072]

[0073] in , , .

[0074] Gain Normalization makes the filter gain 1 in the DC or low-frequency range.

[0075] Therefore, the transfer function of the filter is:

[0076]

[0077] In actual digital implementation, it is also converted into a difference equation through bilinear transformation for real-time operation in DSP or FPGA.

[0078] Step S6: Verify the effect of implementation

[0079] To verify the effectiveness of the present invention, simulation verification and actual engineering verification were carried out.

[0080] (1) Simulation verification of unit universality

[0081] Two doubly-fed induction generator (DFIG) wind power grid-connected system models, 2MW and 7.5kW, were built in the Matlab / Simulink environment. The electrical parameters of the two units differ significantly: the rotor resistance of the 2MW unit is approximately 0.016Ω, and the rotor inductance is approximately 0.08mH; the rotor resistance of the 7.5kW unit is approximately 0.18Ω, and the rotor inductance is approximately 2.5mH. Virtual impedance models using the same design were also constructed. (First-order high-pass filter) and the same oscillation suppressor parameter.

[0082] like Figures 5a-5b As shown, for both different generator sets, the method of this invention can successfully reshape the output impedance curve, ensuring a sufficient amplitude difference between it and the grid impedance curve across the entire frequency band, and significantly improving the phase margin from nearly -15° to over 34°. In comparison, Figures 6a-6b The results show that the traditional virtual impedance method is barely effective on 2MW units, but completely fails on 7.5kW units, with severe resonance spikes appearing in the impedance curve.

[0083] (2) Broadband adaptability verification

[0084] Change the parallel compensation capacitor of the power grid The values ​​of were set to 3μF, 5μF, 8μF, and 11μF, respectively. With different capacitance values, the original resonant frequency of the system drifted from approximately 800Hz to approximately 1100Hz. For example... Figure 7 As shown, after using the method of this invention, the phase margin of the reshaped system remains stable, all within the range of 34° to 38°, which is much higher than the stability margin requirement (usually not less than 30°). This indicates that the present invention can automatically adapt to changes in the power grid structure without any parameter adjustment.

[0085] (3) Verification in actual engineering

[0086] The algorithm of this invention was embedded in the TMS320F28335 DSP of an actual wind farm converter in C code form, increasing the code size by approximately 2KB without adding any additional hardware. This wind farm is connected to... Figure 10 The 220kV system shown here experienced frequent high-frequency oscillations around 864Hz at node 24.

[0087] Figure 11a The voltage waveform at the common coupling point is shown without the suppressor in operation; the voltage is severely distorted, with a THD as high as 56.13%. After the suppressor of this invention is applied at t=1.3s, as shown... Figure 11b As shown, the voltage waveform quickly recovered to a standard sine wave within about 60ms, the THD dropped to 1.35%, the current THD dropped to 2.64%, and the system returned to stable operation.

[0088] Example 2

[0089] This embodiment provides an apparatus for implementing the method described in Embodiment 1.

[0090] The device includes:

[0091] Data acquisition unit: Includes voltage transformers and current transformers, used to acquire stator three-phase voltage, stator three-phase current, rotor three-phase current, common coupling point voltage, and grid-side voltage signals in real time. The sampling frequency is not less than 10kHz, and the data is sent to the control unit after anti-aliasing filtering.

[0092] Grey relational analysis module (optional): This module can run online or offline. When running online, it continuously updates the system impedance model using frequency sweep data from the most recent period, periodically performs variable-weight grey relational analysis, and outputs the identifiers of the dominant influencing factors under the current operating conditions. These identifiers can be used to diagnose system stability trends, but are not a core essential feature of this invention.

[0093] Controller reconfiguration module: This is the core functional module, implementing steps S1-S3 in the digital control program. Specifically, it includes:

[0094] Read the instantaneous value of stator voltage ;

[0095] The stator voltage signal is sent into the control loops of GSC and RSC respectively;

[0096] Set the feedforward injection point in the digital control delay module. Previously, it was added before the last calculation cycle before the pulse width modulation signal was generated;

[0097] By inserting a uniform virtual impedance digital filter into the feedforward path, the following can be achieved: Discretized form;

[0098] Optionally, a series oscillation suppressor The digital implementation.

[0099] Impedance reshaping module: the core of the algorithm logic, which executes... Real-time calculations conceptually enable online reshaping of system impedance. In actual code, no explicit calculations are required. Instead, it achieves the equivalent effect automatically through direct computation of the feedforward path.

[0100] Filtering and compensation module: used to implement the phase compensator and second-order Chebyshev filter A digital filtering algorithm is proposed. It is implemented using a direct type II architecture to save storage space and improve numerical stability.

[0101] Signal synthesis and output module: The signal after feedforward compensation is superimposed with the output signal of the original current loop PI controller to generate a modified modulation wave, which is then converted into a switching signal by the space vector pulse width modulation module and output to the power switching device (IGBT or SiC MOSFET) of the converter.

[0102] All modules of this device can be integrated into the existing digital signal processor (DSP, such as TMS320F28379D) or field programmable gate array (FPGA, such as Xilinx Zynq-7000) of the doubly fed wind turbine converter without increasing the additional hardware cost, and the sampling period of all digital filtering algorithms is consistent with the original control period (usually 100μs or 50μs).

[0103] Example 3

[0104] This embodiment provides a doubly fed wind turbine converter control system that includes the above-described device.

[0105] The system includes:

[0106] Power main circuit: grid-side converter (three-phase full bridge), rotor-side converter (three-phase full bridge), DC bus capacitor, LCL filter (optional);

[0107] Control circuit: includes a doubly fed fan high-frequency oscillation suppression device with stator voltage feedforward as described in Example 2;

[0108] Drive circuit: Isolates and amplifies the pulse width modulation signal output from the control circuit to drive the power switching device.

[0109] This control system can directly replace the control software of existing wind farm converters without modifying the main circuit hardware of the unit, achieving a seamless upgrade of the high-frequency oscillation suppression function.

[0110] Those skilled in the art should understand that the specific values ​​given in the above embodiments (such as 500Hz, 1500Hz, 1200Hz, 0.3, etc.) are all exemplary preferred values ​​and not limitations on the present invention. The core of the present invention lies in the control structure and impedance reshaping expression. In practical applications, those skilled in the art can reasonably select [the appropriate values] based on the actual parameters of the specific system, such as the switching frequency, control delay, and oscillation frequency band. The forms and coefficients, the cutoff frequency and order of the phase compensator, and the cutoff frequency and type of the filter (Chebyshev, Butterworth, Bessel, etc.) are all variations within the scope of protection of this invention.

[0111] Furthermore, the virtual impedance coefficient of the present invention The design can be implemented either by discretizing the continuous-domain transfer function or by directly designing it as a digital filter in the discrete domain, such as using a finite impulse response (FIR) filter or an infinite impulse response (IOR) filter. Any implementation that satisfies the structural characteristics of unified two-sided design and a shifted feedforward point is considered an equivalent embodiment of this invention.

[0112] The method and apparatus provided by this invention can be widely applied to wind farms containing doubly-fed induction generators, weak AC power grids with parallel compensation capacitors, and various new energy grid-connected systems that require suppression of high-frequency oscillations. Implemented through a pure software upgrade, requiring no hardware modifications, it offers significant economic benefits and promising prospects for widespread application.

[0113] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for suppressing high-frequency oscillations in a doubly-fed wind turbine with stator voltage feedforward, characterized in that, Includes the following steps: Step S1: Simultaneously introduce a stator voltage feedforward channel into the control loops of the grid-side converter GSC and the rotor-side converter RSC. Step S2: Set the feedforward point of the stator voltage feedforward channel in the digital control delay stage. Before; Step S3: In the stator voltage feedforward channel, configure a unified virtual impedance coefficient for the grid-side converter and the rotor-side converter. That is, satisfying ,in The virtual impedance transfer function on the GSC side. The virtual impedance transfer function on the RSC side; Step S4: Using the stator voltage feedforward and the unified virtual impedance coefficient The overall output impedance of the doubly fed wind turbine grid-connected system is reshaped to The form; in, For impedance reshaping equivalent coefficient, The output impedance before reshaping. This is a digital control delay transfer function.

2. The method according to claim 1, characterized in that, The virtual impedance coefficient It can take one of the following forms: proportionality coefficient ; First-order high-pass filter ,in The cutoff angular frequency; First-order low-pass filter ,in The cutoff angular frequency; Proportional-differential components .

3. The method according to claim 1, characterized in that, The process includes a preparatory step S0 before step S1: An impedance model of the doubly fed wind turbine grid-connected system was established based on impedance analysis. Variable-weight grey relational analysis is used to quantify control delay. , and add capacitor The sensitivity of multiple influencing factors, including those related to the change in system resonant frequency, was studied, and strong correlation factors that play a dominant role in high-frequency oscillation were identified. The variable-weight grey relational analysis includes: Define sensitivity index ,in , These are the actual and reference values ​​of the resonant frequency, respectively. , These are the actual and baseline values ​​of the influencing factors, respectively. With the aforementioned sensitivity index Using the weights, calculate the grey relational degree of each influencing factor, and determine the strongly correlated factors based on the magnitude of the relational degree.

4. The method according to claim 1, characterized in that, The method further includes step S5: An oscillation suppressor is connected in series in the stator voltage feedforward channel. The oscillation suppressor consists of a phase compensator. and filter Cascaded configuration; The phase compensator Used to improve the phase margin of the system in the high-frequency band, its transfer function is: ,in The first corner frequency, The second corner frequency, and , For the advanced order, The value range is 1 to 3; The filter It is used to suppress high-frequency noise and is a second-order Chebyshev filter or a second-order Butterworth filter.

5. The method according to claim 4, characterized in that, The phase compensator The parameter configuration must satisfy at least one of the following: , ,and ; The advanced order The value is adaptively determined based on the system's phase margin requirements, and its value ensures that the compensated phase margin is greater than or equal to 30°. When the phase compensator is implemented digitally, it is converted into a difference equation through a bilinear transformation.

6. The method according to claim 4, characterized in that, The filter For a second-order Chebyshev filter, its design parameters satisfy: Passband ripple factor Cutoff frequency Set to 1.2 to 1.5 times the upper limit of the oscillation frequency band; Its transfer function is Among them, the extreme points , , Depend on and Determined, gain Normalize the passband gain to 1.

7. The method according to claim 1, characterized in that, The impedance reshaping equivalent coefficient This makes the reshaped output impedance Rotor resistance independent of doubly fed wind turbine Rotor inductance Stator resistance Stator inductor and mutual induction Any specific value in the parameters can be used to achieve the same set of suppressor parameters. It can be directly adapted to doubly fed wind turbines of different electrical parameters without the need for readjustment.

8. A high-frequency oscillation suppression device for a doubly-fed wind turbine with stator voltage feedforward, characterized in that, include: The controller reconfiguration module is used to simultaneously introduce stator voltage feedforward channels into the control loops of the grid-side converter GSC and the rotor-side converter RSC, and to set the feedforward point in the digital control delay stage. Previously, a uniform virtual impedance coefficient was configured for both converters. ; Impedance reshaping module, used to reshape the stator voltage feedforward and the unified virtual impedance coefficient. The overall output impedance of the doubly fed wind turbine grid-connected system is reshaped online to... In the form of, ; The controller reconfiguration module and impedance reshaping module are integrated into a digital signal processor (DSP) or a field-programmable gate array (FPGA).

9. The apparatus according to claim 8, characterized in that, It also includes at least one of the following modules: The grey relational analysis module is used to quantify the sensitivity of multiple influencing factors to changes in the system's resonant frequency based on variable weight grey relational analysis, and output the identifiers of the strongly correlated factors. The filtering and compensation module includes a cascaded phase compensator and a second-order Chebyshev filter. The transfer function of the phase compensator is... The cutoff frequency of the second-order Chebyshev filter is 1200Hz; The data acquisition unit is used to acquire stator voltage, stator current, common coupling point voltage, and grid-side voltage signals in real time.

10. A doubly-fed wind turbine converter control system, characterized in that, include: The main power circuit includes a grid-side converter, a rotor-side converter, and a DC bus capacitor. The control circuit includes a stator voltage feedforward high-frequency oscillation suppression device for doubly fed wind turbines as described in claim 8 or 9. The control circuit outputs a pulse width modulation signal to the power switching devices of the grid-side converter and the rotor-side converter.

Images