Direct-drive fan grid-connected inverter high-frequency resonance dynamic suppression method and system

Abstract

The invention relates to the technical field of resonance suppression, in particular to a direct-drive fan grid-connected inverter high-frequency resonance dynamic suppression method and system, and the method comprises the steps: obtaining a preliminary damping value according to extracted high-frequency resonance natural frequency oscillation distribution data; based on the initial damping value and a power grid voltage fluctuation influence factor, performing dynamic compensation adjustment on the active resonance suppression time delay factor to obtain a high-frequency resonance characteristic value; calculating a frequency drift offset according to the high-frequency resonance characteristic value and a bandwidth upper limit value of a grid-connected inverter controller; and when the frequency drift offset exceeds a preset drift threshold range, optimizing the active damping gain parameter by taking minimization of a resonance peak value and maximization of a dynamic response margin as optimization targets, outputting a high-frequency resonance active suppression parameter optimization set, and completing high-frequency resonance active suppression of the direct-drive fan grid-connected inverter system. By optimizing the active damping gain parameters, high-frequency resonance of the direct-drive fan grid-connected inverter system is effectively suppressed, and stable operation of the system is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of resonance suppression, and in particular to a method and system for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter. Background Art

[0002] With the rapid development of renewable energy power generation technology, direct-drive permanent magnet synchronous wind power generation systems have been widely used in the field of wind power grid connection due to their advantages such as simple structure, high efficiency and low maintenance cost. However, direct-drive wind turbine grid-connected inverter systems face the problem of high-frequency resonance in actual applications, which has a significant impact on the stability of the system. With the large-scale grid connection of wind power generation, in direct-drive wind turbine grid-connected inverter systems, due to the rapid switching action of power electronic devices and the complex interaction between the grid and the inverter, high-frequency resonance occurs frequently. This high-frequency resonance not only causes serious distortion of the grid-side current, thereby affecting the power quality, but also distorts the inverter output voltage waveform, increasing equipment loss and failure risk. Especially when the grid conditions are complex or the load changes drastically, the high-frequency resonance problem is more prominent, seriously threatening the safe and stable operation of the power grid.

[0003] At present, although some suppression methods have been proposed for the high-frequency resonance problem in the direct-drive wind turbine grid-connected inverter system, these methods still have many shortcomings in practical applications. On the one hand, the natural frequency of high-frequency resonance is affected by the grid impedance, filter parameters and inverter control characteristics, and presents a complex distribution characteristic. Traditional resonance suppression methods are usually based on fixed filter parameter design, which is difficult to adapt to the dynamic changes of grid impedance, resulting in poor resonance suppression effect. Especially under weak grid conditions, the wide range of grid impedance changes will further aggravate the drift of the resonant frequency, making it impossible for the active damping control with fixed parameters to effectively cover the resonant frequency band, thereby causing system oscillation or resonance. On the other hand, the existing active damping strategy is insufficient in terms of dynamic response margin. Since the grid-connected inverter needs to meet the dual requirements of smooth power output and high-frequency resonance suppression, the selection of the controller bandwidth and damping gain often needs to be weighed. If the damping gain is set too high, although it can enhance the resonance suppression effect to a certain extent, it may cause the system dynamic response to slow down and affect the rapidity of power regulation; if the damping gain is set too low, the resonance cannot be effectively suppressed, resulting in serious distortion of the grid-side current. In addition, the fluctuation of the grid voltage will further affect the stability of the resonance suppression, making it difficult for traditional control methods to maintain good performance under complex working conditions.

[0004] Therefore, the existing technology has many defects in solving the high-frequency resonance problem of direct-drive wind turbine grid-connected inverters. It is urgent to develop a more efficient and reliable high-frequency resonance active suppression method, which is of great significance for improving the stability and reliability of direct-drive wind turbine grid-connected inverter systems and ensuring the reliable operation of direct-drive wind turbine grid-connected systems under high-frequency resonance conditions. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a method and system for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter.

[0006] In a first aspect, the present invention provides a method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter, the method comprising the following steps: Extracting high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data; Determining an active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtaining a preliminary damping value under an equivalent series resistance effect according to the active resonance suppression delay factor and a grid-connected power smoothing constraint condition; Superimposing the preliminary damping value with the pre-acquired grid voltage fluctuation influence factor to obtain a superposition result, and dynamically compensating and adjusting the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value; Calculating a frequency drift offset according to a difference between the high-frequency resonance characteristic value and a bandwidth upper limit value of the grid-connected inverter controller; When the frequency drift offset exceeds a preset drift threshold range, optimizing the active damping gain parameters with minimizing the resonance peak and maximizing the dynamic response margin as optimization goals, and outputting an optimized set of high-frequency resonance active suppression parameters; The high-frequency resonance active suppression parameter optimization set is used to perform high-frequency resonance active suppression on a direct-drive wind turbine grid-connected inverter system.

[0007] In a further embodiment, the step of extracting high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data includes: Collect grid-side current data and grid-connected inverter output voltage waveform data, and use a sliding window mean filter algorithm to remove high-frequency noise from the grid-side current data to obtain grid-side denoised current data; performing amplitude normalization processing on the output voltage waveform data of the grid-connected inverter to obtain standardized voltage waveform data; Performing spectrum analysis on the grid-side de-noised current data using a fast Fourier transform to obtain a frequency domain amplitude spectrum, and extracting distribution data of current distortion frequency components of each harmonic component and a high-frequency band based on the amplitudes of the frequency components in the frequency domain amplitude spectrum; Performing waveform analysis on the standardized voltage waveform data to extract voltage oscillation characteristic parameters; When it is determined that high-frequency resonance occurs, the high-frequency resonance natural frequency oscillation distribution data is determined based on the current distortion frequency component distribution data and the voltage oscillation characteristic parameters in the high-frequency resonance coincidence frequency band.

[0008] In a further embodiment, the step of determining the active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data includes: Obtaining measured reactance parameters of a filter in a direct-drive wind turbine grid-connected inverter system, and determining a main resonant frequency point based on the high-frequency resonant natural frequency oscillation distribution data; Extracting equivalent reactance distribution data of the filter at the main resonant frequency point according to the measured reactance parameters of the filter; Based on the equivalent reactance distribution data, fitting the reactance-frequency relationship by the least squares method to generate a reactance-frequency characteristic model; Determining the impedance mutation degree of the high-frequency resonant frequency band according to the impedance change rate of the reactance-frequency characteristic model, and determining the impedance mutation frequency band range according to the impedance mutation degree and a preset mutation threshold; In the impedance mutation frequency range, using the main resonant frequency point to calculate the initial value of the active damping gain under the phase compensation angle constraint; It is determined whether the initial value of the active damping gain satisfies a preset phase margin stability condition. If so, an active resonance suppression delay factor is obtained based on the initial value of the active damping gain using a phase compensation relationship.

[0009] In a further embodiment, the step of obtaining a preliminary damping value under the equivalent series resistance effect according to the active resonance suppression delay factor and the grid-connected power smoothing constraint condition includes: Obtaining a grid-connected power smoothing constraint condition of a direct-drive wind turbine grid-connected inverter system, quantifying the grid-connected power smoothing constraint condition into a frequency-domain-time-domain joint indicator, and generating a grid-connected power smoothing constraint parameter set; Obtaining an equivalent series resistance according to the active resonance suppression delay factor and the main resonance frequency point; Inputting the equivalent series resistance into the closed-loop transfer function of the direct-drive wind turbine grid-connected inverter system to calculate the corresponding grid-side current harmonic distortion rate and active power fluctuation rate; comparing the grid-side current harmonic distortion rate and the active power fluctuation rate with the grid-connected power smoothing constraint parameter set, and extracting an active damping gain value corresponding to a current equivalent series resistance when both the grid-side current harmonic distortion rate and the active power fluctuation rate satisfy the grid-connected power smoothing constraint parameter set; The active damping gain value corresponding to the current equivalent series resistance is used as the preliminary damping value under the equivalent series resistance effect.

[0010] In a further embodiment, the grid-connected power smoothing constraint condition includes at least an active power fluctuation rate constraint condition and a total harmonic distortion rate constraint condition.

[0011] In a further embodiment, the step of dynamically compensating and adjusting the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value includes: According to the superposition result, adjusting the active resonance suppression delay factor through a linear compensation relationship to obtain a corrected value of the active resonance suppression delay factor; Using the active resonance suppression delay factor correction value to adjust the delay compensation amount of the pulse modulation signal in the grid-connected inverter controller in real time; Delay compensation closed-loop control is performed according to the delay compensation amount, grid-side current time series data after closed-loop control is obtained, and spectrum analysis is performed on the grid-side current time series data to extract high-frequency resonance characteristic values.

[0012] In a further embodiment, the high-frequency resonance characteristic value includes the resonance frequency point, resonance amplitude and phase mutation interval obtained by dynamic compensation after superimposing the grid fluctuation.

[0013] In a further embodiment, the step of optimizing the active damping gain parameters with minimizing the resonance peak and maximizing the dynamic response margin as the optimization goal and outputting the optimized set of high-frequency resonance active suppression parameters includes: Define the active damping gain parameter as a decision variable and set the optimization range of the active damping gain parameter; With the optimization goals of minimizing the resonance peak and maximizing the dynamic response margin, several particles are randomly generated within the optimization range of the active damping gain parameter to form an initial particle swarm. The active damping gain parameter of each particle is input into the grid-connected inverter controller. The resonance peak is obtained through closed-loop simulation. The phase margin and amplitude margin under the current active damping gain parameter are calculated to obtain the dynamic response margin. Calculating a current fitness value of each particle according to the resonance peak value and the dynamic response margin, and updating an individual extreme value position according to the current fitness value of each particle; Finding a global optimal solution among the individual extreme value positions of all particles, updating the global extreme value position according to the global optimal solution, and updating the speed and position of each particle according to the updated individual extreme value position and the global extreme value position; Repeat the above steps until the fitness value converges, and output the active damping gain parameter combination corresponding to the global optimal extreme value position to form the optimized set of high-frequency resonance active suppression parameters.

[0014] In a further embodiment, the fitness function of each particle is defined as a weighted sum of the inverse of the resonance peak and the dynamic response margin.

[0015] In a second aspect, the present invention provides a system for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter, the system comprising: A data acquisition module is used to extract high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data; a damping analysis module, configured to determine an active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtain a preliminary damping value under an equivalent series resistance effect according to the active resonance suppression delay factor and a grid-connected power smoothing constraint condition; a dynamic compensation module, configured to superimpose the preliminary damping value with a pre-acquired grid voltage fluctuation influence factor to obtain a superposition result, and dynamically compensate and adjust the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value; a drift analysis module, configured to calculate a frequency drift offset based on a difference between the high-frequency resonance characteristic value and a bandwidth upper limit value of the grid-connected inverter controller; a parameter optimization module for optimizing the active damping gain parameters with the optimization objectives of minimizing the resonance peak and maximizing the dynamic response margin when the frequency drift offset exceeds a preset drift threshold range, and outputting an optimized set of high-frequency resonance active suppression parameters; The resonance suppression module is used to perform high-frequency resonance active suppression on a direct-drive wind turbine grid-connected inverter system using the high-frequency resonance active suppression parameter optimization set.

[0016] The present invention provides a method and system for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter. The method extracts high-frequency resonance natural frequency oscillation distribution data from collected grid-side current data and grid-connected inverter output voltage waveform data; determines an active resonance suppression time delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtains a preliminary damping value under the equivalent series resistance effect according to the active resonance suppression time delay factor and a grid-connected power smoothing constraint condition; superimposes the preliminary damping value with a pre-acquired grid voltage fluctuation influencing factor to obtain a superposition result, and dynamically compensates and adjusts the active resonance suppression time delay factor according to the superposition result to obtain a high-frequency resonance characteristic value; calculates a frequency drift offset according to a difference between the high-frequency resonance characteristic value and a bandwidth upper limit value of a grid-connected inverter controller; when the frequency drift offset exceeds a preset drift threshold range, optimizes an active damping gain parameter with minimizing the resonance peak and maximizing the dynamic response margin as optimization goals, and outputs an optimized set of high-frequency resonance active suppression parameters; and uses the optimized set of high-frequency resonance active suppression parameters to actively suppress high-frequency resonance of the direct-drive wind turbine grid-connected inverter system. Compared with the existing technology, this method optimizes the active damping gain parameters to effectively suppress high-frequency resonance in the direct-drive wind turbine grid-connected inverter system under complex working conditions, improves the overall operating performance of the system, and ensures safe and stable operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter provided by an embodiment of the present invention; Figure 2 This is a block diagram of a high-frequency resonance dynamic suppression system for a direct-drive wind turbine grid-connected inverter provided by an embodiment of the present invention.

[0018] Explanation of reference numerals: 101, data acquisition module; 102, damping analysis module; 103, dynamic compensation module; 104, drift analysis module; 105, parameter optimization module; 106, resonance suppression module. DETAILED DESCRIPTION

[0019] The following describes the embodiments of the present invention in detail with reference to the accompanying drawings. The embodiments are provided for illustrative purposes only and are not to be construed as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of protection of the present invention. Many changes may be made to the present invention without departing from the spirit and scope of the present invention.

[0020] refer to Figure 1 The embodiment of the present invention provides a method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter, such as Figure 1 As shown, the method includes the following steps: S1. Extract high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data.

[0021] In some embodiments, the step of extracting high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data includes: Collect grid-side current data and grid-connected inverter output voltage waveform data, and use a sliding window mean filter algorithm to remove high-frequency noise from the grid-side current data to obtain grid-side denoised current data; performing amplitude normalization processing on the output voltage waveform data of the grid-connected inverter to obtain standardized voltage waveform data; Performing spectrum analysis on the grid-side de-noised current data using a fast Fourier transform to obtain a frequency domain amplitude spectrum, and extracting distribution data of current distortion frequency components of each harmonic component and a high-frequency band based on the amplitudes of the frequency components in the frequency domain amplitude spectrum; Performing waveform analysis on the standardized voltage waveform data to extract voltage oscillation characteristic parameters; When it is determined that high-frequency resonance occurs, the high-frequency resonance natural frequency oscillation distribution data is determined based on the current distortion frequency component distribution data and the voltage oscillation characteristic parameters in the high-frequency resonance coincidence frequency band.

[0022] Specifically, this embodiment collects grid-side current data from the grid connection point of the direct-drive wind turbine grid-connected inverter system (i.e., the interface between the grid-connected inverter output and the grid), and then removes high-frequency noise from the grid-side current data based on a sliding window mean filtering algorithm. For example, high-frequency switching noise above 10 kHz is removed, and grid-side denoised current data with valid frequency band signals is retained. The grid-side denoised current data is then subjected to a fast Fourier transform to calculate its frequency domain amplitude spectrum. This embodiment extracts harmonic components (such as the 5th, 7th, and 11th harmonics) and high-frequency distortion frequency components based on the percentage of the amplitude of the frequency component relative to the fundamental wave amplitude in the frequency domain amplitude spectrum to generate current distortion frequency component distribution data. The current distortion frequency component distribution data may include, but is not limited to, frequency points, corresponding amplitude percentages, and phase angle data. At the same time, this embodiment collects grid-connected inverter output voltage waveform data from the output end of the DC-AC conversion unit of the grid-connected inverter and performs amplitude normalization processing on the grid-connected inverter output voltage waveform data to eliminate DC offset and generate standardized voltage waveform data. For the standardized voltage waveform data, this embodiment uses a zero-crossing detection algorithm to calculate the voltage waveform oscillation frequency. Based on a peak detection algorithm, this embodiment extracts the voltage waveform oscillation amplitude and the oscillation attenuation rate of adjacent peaks (adjacent peak attenuation ratio) to generate voltage oscillation characteristic parameters. The voltage oscillation characteristic parameters may include oscillation frequency, amplitude, attenuation rate, and time-domain waveform distortion rate. If the current distortion frequency component distribution data contains a component with a frequency greater than 1 kHz and its amplitude exceeds a preset amplitude threshold (e.g., 3% of the fundamental wave amplitude), it is marked as a high-frequency distortion frequency band. This embodiment uses the oscillation frequency in the voltage oscillation characteristic parameters to determine whether the high-frequency distortion frequency band overlaps with the voltage oscillation frequency. If the oscillation attenuation rate of the overlapping frequency band is less than a preset oscillation attenuation threshold (i.e., the oscillation continues without attenuation), it is determined that high-frequency resonance has occurred, and a high-frequency resonance occurrence flag signal and the corresponding resonance overlap frequency band range are output.

[0023] Within the resonant overlap frequency band, this embodiment extracts frequency points, amplitude, and phase data of the high-frequency distortion frequency band marked in the current distortion frequency component distribution data. Combined with the oscillation amplitude and attenuation rate of the same frequency band in the voltage oscillation characteristic parameters, the intersection frequency point is taken to determine the main resonant frequency. Based on the amplitude-phase distribution of the main resonant frequency point, the oscillation amplitude envelope and phase mutation interval of the resonant natural frequency are generated, thereby obtaining a high-frequency resonant natural frequency oscillation distribution dataset. The high-frequency resonant natural frequency oscillation distribution dataset may include, but is not limited to, the main resonant frequency point, the resonant amplitude envelope (a distribution curve showing amplitude variation with frequency), the phase mutation interval, and the resonant modal energy ratio. The phase mutation interval is the frequency range in which the phase angle jump exceeds 45°, and the resonant modal energy ratio is the percentage of high-frequency distortion energy to total distortion energy.

[0024] S2. Determine an active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtain a preliminary damping value under the equivalent series resistance effect according to the active resonance suppression delay factor and the grid-connected power smoothing constraint condition.

[0025] In some embodiments, the step of determining the active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data includes: Obtaining measured reactance parameters of a filter in a direct-drive wind turbine grid-connected inverter system, and determining a main resonant frequency point based on the high-frequency resonant natural frequency oscillation distribution data; Extracting equivalent reactance distribution data of the filter at the main resonant frequency point according to the measured reactance parameters of the filter; Based on the equivalent reactance distribution data, fitting the reactance-frequency relationship by the least squares method to generate a reactance-frequency characteristic model; Determining the impedance mutation degree of the high-frequency resonant frequency band according to the impedance change rate of the reactance-frequency characteristic model, and determining the impedance mutation frequency band range according to the impedance mutation degree and a preset mutation threshold; In the impedance mutation frequency range, using the main resonant frequency point to calculate the initial value of the active damping gain under the phase compensation angle constraint; It is determined whether the initial value of the active damping gain satisfies a preset phase margin stability condition. If so, an active resonance suppression delay factor is obtained based on the initial value of the active damping gain using a phase compensation relationship.

[0026] Specifically, this embodiment obtains the measured reactance parameters of the inductor-capacitor-inductor filter (i.e., the LCL filter located between the output end of the grid-connected inverter and the power grid) configured in the direct-drive wind turbine grid-connected inverter based on the main resonant frequency point in the high-frequency resonant natural frequency oscillation distribution data set. For example, the measured reactance parameters of the filter may include the machine-side inductance value, the machine-side capacitance value, and the grid-side inductance value. This embodiment combines the measured reactance parameters of the inductor-capacitor-inductor filter to calculate the equivalent inductive reactance and capacitive reactance components of the filter at the main resonant frequency point, and generates the equivalent reactance distribution data of the filter in the high-frequency resonant frequency band. Based on the equivalent reactance distribution data, the reactance-frequency relationship is polynomially fitted by the least squares method to generate a reactance-frequency characteristic model. The reactance-frequency characteristic model characterizes the filter in the resonant frequency band. The impedance characteristic reflects how the filter impedance changes with frequency. This embodiment determines the slope of the impedance change with frequency within the high-frequency resonant frequency band based on the first-order derivative of the reactance-frequency characteristic model (i.e., the impedance change rate). The degree of impedance mutation in the high-frequency resonant frequency band is determined based on the impedance change rate. If the impedance change rate exceeds a preset mutation threshold (e.g., 10Ω / Hz), an impedance mutation is determined, requiring phase compensation. In this case, an impedance mutation flag signal and the impedance mutation frequency band range are output. Next, based on the impedance mutation frequency band range and the phase compensation relationship (the product of the delay factor and the resonant frequency must satisfy the phase compensation angle constraint, which can be set to 90°), this embodiment calculates the initial value of the active damping gain that satisfies the phase compensation angle constraint. The specific calculation formula for the initial value of the active damping gain is: Where, is the initial value of the active damping gain, unit is dimensionless; is the radian value corresponding to the phase compensation angle constraint value; is the preset empirical delay factor, in seconds; is the main resonant frequency.

[0027] Next, this embodiment verifies whether the initial value of the active damping gain satisfies the phase margin stability condition of not less than 45° based on the pole distribution of the system open-loop transfer function. If so, the active resonance suppression delay factor is output by inversely deducing the phase compensation relationship based on the initial value of the active damping gain. Specifically, this embodiment substitutes the initial value of the active damping gain into the system open-loop transfer function to calculate the corresponding phase margin. If the phase margin is not less than 45°, the active resonance suppression delay factor is inversely deduced based on the phase compensation relationship. The specific calculation formula is: Where, is the active resonance suppression delay factor; is the radian value corresponding to the actual phase compensation angle.

[0028] If the initial value of the active damping gain does not meet the phase margin stability condition of a phase margin of not less than 45°, the high-frequency resonance natural frequency oscillation distribution data is reacquired, and the active resonance suppression delay factor is determined by the high-frequency resonance natural frequency oscillation distribution data. It should be noted that active damping control suppresses resonance by introducing a virtual impedance (equivalent series resistance), and its effect is equivalent to adding phase compensation to the system; the phase compensation angle is defined as the phase lag introduced by the virtual impedance that needs to offset the phase mutation caused by the resonance.

[0029] In some embodiments, the step of obtaining a preliminary damping value under the equivalent series resistance effect according to the active resonance suppression delay factor and the grid-connected power smoothing constraint condition includes: Obtaining a grid-connected power smoothing constraint condition of a direct-drive wind turbine grid-connected inverter system, quantifying the grid-connected power smoothing constraint condition into a frequency-domain-time-domain joint indicator, and generating a grid-connected power smoothing constraint parameter set; Obtaining an equivalent series resistance according to the active resonance suppression delay factor and the main resonance frequency point; Inputting the equivalent series resistance into the closed-loop transfer function of the direct-drive wind turbine grid-connected inverter system to calculate the corresponding grid-side current harmonic distortion rate and active power fluctuation rate; comparing the grid-side current harmonic distortion rate and the active power fluctuation rate with the grid-connected power smoothing constraint parameter set, and extracting an active damping gain value corresponding to a current equivalent series resistance when both the grid-side current harmonic distortion rate and the active power fluctuation rate satisfy the grid-connected power smoothing constraint parameter set; The active damping gain value corresponding to the current equivalent series resistance is used as the preliminary damping value under the equivalent series resistance effect.

[0030] Specifically, this embodiment obtains the grid-connected power smoothing constraint conditions of the direct-drive wind turbine grid-connected inverter system based on the power smoothing constraint conditions that the direct-drive wind turbine grid-connected inverter system needs to meet during operation to ensure system stability and power quality. The grid-connected power smoothing constraint conditions may include active power fluctuation constraint conditions, total harmonic distortion rate constraint conditions, and dynamic response time constraint conditions. For example, the active power fluctuation rate constraint condition may be that the percentage change of grid-connected active power in adjacent cycles must be no more than 2% of the rated power, the total harmonic distortion rate constraint condition may be that the grid-side current harmonic distortion rate must be no more than 5%, and the dynamic response time constraint condition may be that the time from disturbance to recovery of stability of the system must be less than 100ms. This embodiment quantifies the above-mentioned grid-connected power smoothing constraint conditions into a frequency domain-time domain joint indicator. For example, the above-mentioned grid-connected power smoothing constraint conditions are quantified into an active power fluctuation rate threshold limit and a total harmonic distortion rate threshold limit to generate a grid-connected power smoothing constraint parameter set. According to the active damping principle, this embodiment uses virtual impedance to be equivalent to a series resistance effect. The specific calculation formula of the equivalent series resistance value is: Where, is the equivalent series resistance value; is the active damping gain, and the initial value of the active damping gain can be set to 1; is the active resonance suppression delay factor; is the main resonant frequency.

[0031] In this embodiment, active damping is equivalent to a series resistance effect through virtual impedance, but it is necessary to ensure that the virtual impedance value provides sufficient damping effect within the resonant frequency band while not deteriorating the dynamic response of the system. In this embodiment, the equivalent series resistance value is substituted into the closed-loop transfer function of the direct-drive wind turbine grid-connected inverter system to obtain the frequency response characteristics of the system. By analyzing the frequency response characteristics, the corresponding grid-side current harmonic distortion rate is calculated, and the percentage of power change in adjacent cycles is calculated through time domain simulation to obtain the active power fluctuation rate. Assuming that the open-loop transfer function of the system is G(s), the closed-loop transfer function after adding the equivalent series resistance value is: Where, is the closed-loop transfer function of the direct-drive wind turbine grid-connected inverter system.

[0032] In this embodiment, the grid-side current harmonic distortion rate and the active power fluctuation rate are respectively compared with the grid-connected power smoothing constraint parameter set. When the grid-side current harmonic distortion rate and the active power fluctuation rate both satisfy the grid-connected power smoothing constraint parameter set, the active damping gain value corresponding to the current equivalent series resistance is extracted. For example, if the grid-side current harmonic distortion rate is not greater than 5% and the active power fluctuation rate is not greater than 2%, the active damping gain value corresponding to the current equivalent series resistance is extracted as a preliminary damping value under the equivalent series resistance effect. If either the grid-side current harmonic distortion rate or the active power fluctuation rate does not satisfy the grid-connected power smoothing constraint parameter set, the active damping gain value is adjusted using a gradient descent method until the grid-connected power smoothing constraint parameter set is satisfied.

[0033] S3. Superimposing the preliminary damping value with the pre-acquired grid voltage fluctuation influence factor to obtain a superposition result, and dynamically compensating and adjusting the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value.

[0034] In some embodiments, the step of dynamically compensating and adjusting the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value includes: According to the superposition result, adjusting the active resonance suppression delay factor through a linear compensation relationship to obtain a corrected value of the active resonance suppression delay factor; Using the active resonance suppression delay factor correction value to adjust the delay compensation amount of the pulse modulation signal in the grid-connected inverter controller in real time; Delay compensation closed-loop control is performed according to the delay compensation amount, grid-side current time series data after closed-loop control is obtained, and spectrum analysis is performed on the grid-side current time series data to extract high-frequency resonance characteristic values.

[0035] Specifically, this embodiment performs statistical analysis on the grid voltage fluctuation data, calculates the standard deviation and probability distribution of its fluctuation amplitude, generates a grid voltage fluctuation influence factor, and superimposes the preliminary damping value under the equivalent series resistance effect and the grid voltage fluctuation influence factor according to a weight ratio to generate a superposition result. Then, this embodiment adjusts the active resonance suppression delay factor through a linear compensation relationship based on the superposition result. The specific calculation formula can be: Where, is the correction value of active resonance suppression delay factor; is the active resonance suppression delay factor; is a compensation coefficient, which is used to adjust the influence of the superposition result on the active resonance suppression delay factor, and can be set by those skilled in the art based on experience; is the superposition result.

[0036] In this embodiment, the dynamically corrected active resonance suppression delay factor correction value is input into the active damping module of the direct-drive wind turbine grid-connected inverter controller. The active damping module adjusts the delay compensation amount of the pulse modulation signal in real time, suppresses high-frequency resonance through closed-loop control, and collects grid-side current time series data after closed-loop control. The resonant frequency point, resonant amplitude, and phase mutation range obtained by dynamic compensation after superimposing grid fluctuations are extracted from the grid-side current time series data through spectrum analysis. The resonant frequency point is the high-frequency resonant main frequency point in the amplitude spectrum that exceeds 3% of the fundamental wave amplitude, the resonant amplitude is the percentage of the spectrum amplitude relative to the fundamental wave amplitude, and the phase mutation range is the frequency range in which the phase angle jump exceeds 45°, thereby generating a high-frequency resonance characteristic value.

[0037] S4. Calculate the frequency drift offset according to the difference between the high-frequency resonance characteristic value and the bandwidth upper limit of the grid-connected inverter controller.

[0038] S5. When the frequency drift offset exceeds a preset drift threshold range, the active damping gain parameters are optimized with minimizing the resonance peak and maximizing the dynamic response margin as the optimization goals, and an optimized set of high-frequency resonance active suppression parameters is output.

[0039] S6. Use the high-frequency resonance active suppression parameter optimization set to perform high-frequency resonance active suppression on the direct-drive wind turbine grid-connected inverter system.

[0040] In some embodiments, the step of optimizing the active damping gain parameters with minimizing the resonance peak and maximizing the dynamic response margin as the optimization objectives and outputting an optimized set of high-frequency resonance active suppression parameters includes: Define the active damping gain parameter as a decision variable and set the optimization range of the active damping gain parameter; With the optimization goals of minimizing the resonance peak and maximizing the dynamic response margin, several particles are randomly generated within the optimization range of the active damping gain parameter to form an initial particle swarm. The active damping gain parameter of each particle is input into the grid-connected inverter controller. The resonance peak is obtained through closed-loop simulation. The phase margin and amplitude margin under the current active damping gain parameter are calculated to obtain the dynamic response margin. Calculating a current fitness value of each particle according to the resonance peak value and the dynamic response margin, and updating an individual extreme value position according to the current fitness value of each particle; Finding a global optimal solution among the individual extreme value positions of all particles, updating the global extreme value position according to the global optimal solution, and updating the speed and position of each particle according to the updated individual extreme value position and the global extreme value position; Repeat the above steps until the fitness value converges, and output the active damping gain parameter combination corresponding to the global optimal extreme value position to form the optimized set of high-frequency resonance active suppression parameters.

[0041] Specifically, this embodiment sets the optimization range of the active damping gain parameter based on the stability analysis results of the main resonant frequency points and the closed-loop transfer function of the system in the high-frequency resonant natural frequency oscillation distribution data set and according to engineering experience and system constraints (such as the physically achievable range of the damping gain). In the direct-drive wind turbine grid-connected inverter system, the active damping gain parameter is usually achieved by adjusting the parameters of the proportional-integral-differential control algorithm, and the optimization objectives are defined as minimizing the resonant peak and maximizing the dynamic response margin. A random number generation method is used to randomly generate a number of particles within the optimization range of the active damping gain parameter. It is assumed that N=50 particles are generated. Particles, each particle represents a set of active damping gain parameter combinations to form an initial particle group, and then the active damping gain parameters of each particle are input into the grid-connected inverter controller, and the grid-side current spectrum data is generated through closed-loop simulation. The grid-side current spectrum data is subjected to frequency domain analysis, and the maximum amplitude in the high-frequency resonance frequency band is extracted. The percentage relative to the fundamental amplitude is calculated to obtain the maximum resonance peak value, where the resonance peak value refers to the amplitude of the grid-side current at the resonance frequency. At the same time, during the simulation process, this embodiment uses a frequency response analysis method to calculate the phase margin under the current active damping gain parameters based on the system open-loop transfer function model. Phase margin and amplitude margin, where the phase margin refers to the value of the amplitude less than 0dB when the phase angle of the system open-loop frequency response reaches -180°, and the amplitude margin refers to the value of the phase angle greater than -180° when the amplitude of the system open-loop frequency response reaches 0dB. In this embodiment, the smaller value of the phase margin and the amplitude margin is taken as the dynamic response margin. For example, if the phase margin is 48° and the amplitude margin is 7dB, the dynamic response margin is 48°. In this embodiment, the fitness function of each particle is defined as the weighted sum of the inverse of the resonance peak and the dynamic response margin. Then, the current fitness value of the particle is compared with the historical optimal value. The new individual extreme value position is selected, and the global optimal solution is screened from all individual extreme values. The global extreme value position is updated by the global optimal solution, and the particle position is updated according to the particle velocity update formula. When the fitness value change rate for multiple consecutive times is less than the preset convergence threshold, the optimization search is terminated, and the high-frequency resonance active suppression parameter optimization set corresponding to the global optimal extreme value position is output. The harmonic distortion rate and dynamic response margin under this parameter combination are verified through closed-loop simulation. If the verification is passed, the high-frequency resonance active suppression parameter optimization set is output. The high-frequency resonance active suppression parameter optimization set includes the optimized active damping gain parameter and its corresponding resonance peak and phase margin.

[0042] The embodiment of the present invention provides a method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter, the method comprising extracting high-frequency resonance natural frequency oscillation distribution data from collected grid-side current data and grid-connected inverter output voltage waveform data; determining an active resonance suppression time delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtaining a preliminary damping value under the equivalent series resistance effect according to the active resonance suppression time delay factor and the grid-connected power smoothing constraint condition; superimposing the preliminary damping value with a pre-acquired grid voltage fluctuation influencing factor to obtain a superposition result, and According to the superposition result, the active resonance suppression delay factor is dynamically compensated and adjusted to obtain a high-frequency resonance characteristic value; the frequency drift offset is calculated based on the difference between the high-frequency resonance characteristic value and the bandwidth upper limit of the grid-connected inverter controller; when the frequency drift offset exceeds the preset drift threshold range, the active damping gain parameter is optimized with minimizing the resonance peak and maximizing the dynamic response margin as the optimization goals, and an optimized set of high-frequency resonance active suppression parameters is output; the high-frequency resonance active suppression parameter optimized set is used to actively suppress the high-frequency resonance of the direct-drive wind turbine grid-connected inverter system. Compared with the existing technology, this method optimizes the active damping gain parameter to achieve effective suppression of high-frequency resonance in the direct-drive wind turbine grid-connected inverter system under complex working conditions, improve the overall operating performance of the system, and ensure safe and stable operation of the system.

[0043] It should be noted that the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of this application.

[0044] In one embodiment, Figure 2 As shown, an embodiment of the present invention provides a high-frequency resonance dynamic suppression system for a direct-drive wind turbine grid-connected inverter, the system comprising: The data acquisition module 101 is used to extract high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data; a damping analysis module 102 for determining an active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtaining a preliminary damping value under an equivalent series resistance effect according to the active resonance suppression delay factor and a grid-connected power smoothing constraint condition; A dynamic compensation module 103 is configured to superimpose the preliminary damping value with a pre-acquired grid voltage fluctuation influence factor to obtain a superposition result, and dynamically compensate and adjust the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value; a drift analysis module 104 for calculating a frequency drift offset based on a difference between the high-frequency resonance characteristic value and a bandwidth upper limit of the grid-connected inverter controller; a parameter optimization module 105 for optimizing the active damping gain parameters with the optimization objectives of minimizing the resonance peak and maximizing the dynamic response margin when the frequency drift offset exceeds a preset drift threshold range, and outputting an optimized set of high-frequency resonance active suppression parameters; The resonance suppression module 106 is configured to perform high-frequency resonance active suppression on the direct-drive wind turbine grid-connected inverter system using the high-frequency resonance active suppression parameter optimization set.

[0045] For the specific definition of a high-frequency resonance dynamic suppression system for a direct-drive wind turbine grid-connected inverter, please refer to the above-mentioned definition of a high-frequency resonance dynamic suppression method for a direct-drive wind turbine grid-connected inverter, which will not be repeated here. A person of ordinary skill in the art will appreciate that the various modules and steps described in conjunction with the embodiments disclosed in this application can be implemented in hardware, software, or a combination of both. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0046] The embodiment of the present invention provides a high-frequency resonance dynamic suppression system for a direct-drive wind turbine grid-connected inverter. The data acquisition module of the system extracts high-frequency resonance natural frequency oscillation distribution data from the collected grid-side current data and the grid-connected inverter output voltage waveform data; the damping analysis module determines the active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtains a preliminary damping value under the equivalent series resistance effect according to the active resonance suppression delay factor and the grid-connected power smoothing constraint condition; the dynamic compensation module superimposes the preliminary damping value with the pre-acquired grid voltage fluctuation influence factor to obtain a superposition result, and calculates the superposition result according to the dynamic compensation module. The superposition result dynamically compensates and adjusts the active resonance suppression delay factor to obtain a high-frequency resonance characteristic value; the drift analysis module calculates the frequency drift offset based on the difference between the high-frequency resonance characteristic value and the bandwidth upper limit of the grid-connected inverter controller; when the frequency drift offset exceeds the preset drift threshold range, the parameter optimization module optimizes the active damping gain parameters with the optimization goals of minimizing the resonance peak and maximizing the dynamic response margin, and outputs an optimized set of high-frequency resonance active suppression parameters; the resonance suppression module uses the optimized set of high-frequency resonance active suppression parameters to actively suppress the high-frequency resonance of the direct-drive wind turbine grid-connected inverter system. Compared with the existing technology, this system optimizes the active damping gain parameters to achieve effective suppression of high-frequency resonance in the direct-drive wind turbine grid-connected inverter system under complex working conditions, improve the overall operating performance of the system, and ensure safe and stable operation of the system.

[0047] The above-described embodiments merely represent several preferred implementations of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art could make several improvements and substitutions without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be based on the scope of protection of the claims.

Claims

1. A method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter, characterized in that: The following steps are involved: Extracting high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data; Determining an active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtaining a preliminary damping value under an equivalent series resistance effect according to the active resonance suppression delay factor and a grid-connected power smoothing constraint condition; Superimposing the preliminary damping value with the pre-acquired grid voltage fluctuation influence factor to obtain a superposition result, and dynamically compensating and adjusting the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value; Calculating a frequency drift offset according to a difference between the high-frequency resonance characteristic value and a bandwidth upper limit value of the grid-connected inverter controller; When the frequency drift offset exceeds a preset drift threshold range, optimizing the active damping gain parameters with minimizing the resonance peak and maximizing the dynamic response margin as optimization goals, and outputting an optimized set of high-frequency resonance active suppression parameters; The high-frequency resonance active suppression parameter optimization set is used to perform high-frequency resonance active suppression on a direct-drive wind turbine grid-connected inverter system.

2. The method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 1, characterized in that: The step of extracting high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data includes: Collect grid-side current data and grid-connected inverter output voltage waveform data, and use a sliding window mean filter algorithm to remove high-frequency noise from the grid-side current data to obtain grid-side denoised current data; performing amplitude normalization processing on the output voltage waveform data of the grid-connected inverter to obtain standardized voltage waveform data; Performing spectrum analysis on the grid-side de-noised current data using a fast Fourier transform to obtain a frequency domain amplitude spectrum, and extracting distribution data of current distortion frequency components of each harmonic component and a high-frequency band based on the amplitudes of the frequency components in the frequency domain amplitude spectrum; Performing waveform analysis on the standardized voltage waveform data to extract voltage oscillation characteristic parameters; When it is determined that high-frequency resonance occurs, the high-frequency resonance natural frequency oscillation distribution data is determined based on the current distortion frequency component distribution data and the voltage oscillation characteristic parameters in the high-frequency resonance coincidence frequency band.

3. The method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 1, characterized in that: The step of determining the active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data includes: Obtaining measured reactance parameters of a filter in a direct-drive wind turbine grid-connected inverter system, and determining a main resonant frequency point based on the high-frequency resonant natural frequency oscillation distribution data; Extracting equivalent reactance distribution data of the filter at the main resonant frequency point according to the measured reactance parameters of the filter; Based on the equivalent reactance distribution data, fitting the reactance-frequency relationship by the least squares method to generate a reactance-frequency characteristic model; Determining the impedance mutation degree of the high-frequency resonant frequency band according to the impedance change rate of the reactance-frequency characteristic model, and determining the impedance mutation frequency band range according to the impedance mutation degree and a preset mutation threshold; In the impedance mutation frequency range, using the main resonant frequency point to calculate the initial value of the active damping gain under the phase compensation angle constraint; It is determined whether the initial value of the active damping gain satisfies a preset phase margin stability condition. If so, an active resonance suppression delay factor is obtained based on the initial value of the active damping gain using a phase compensation relationship.

4. A method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 3, characterized in that: The step of obtaining a preliminary damping value under the equivalent series resistance effect according to the active resonance suppression delay factor and the grid-connected power smoothing constraint condition includes: Obtaining a grid-connected power smoothing constraint condition of a direct-drive wind turbine grid-connected inverter system, quantifying the grid-connected power smoothing constraint condition into a frequency-domain-time-domain joint indicator, and generating a grid-connected power smoothing constraint parameter set; Obtaining an equivalent series resistance according to the active resonance suppression delay factor and the main resonance frequency point; Inputting the equivalent series resistance into the closed-loop transfer function of the direct-drive wind turbine grid-connected inverter system to calculate the corresponding grid-side current harmonic distortion rate and active power fluctuation rate; comparing the grid-side current harmonic distortion rate and the active power fluctuation rate with the grid-connected power smoothing constraint parameter set, and extracting an active damping gain value corresponding to a current equivalent series resistance when both the grid-side current harmonic distortion rate and the active power fluctuation rate satisfy the grid-connected power smoothing constraint parameter set; The active damping gain value corresponding to the current equivalent series resistance is used as the preliminary damping value under the equivalent series resistance effect.

5. The method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 4, characterized in that: The grid-connected power smoothing constraint condition includes at least an active power fluctuation rate constraint condition and a total harmonic distortion rate constraint condition.

6. The method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 1, characterized in that: The step of dynamically compensating and adjusting the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value includes: According to the superposition result, adjusting the active resonance suppression delay factor through a linear compensation relationship to obtain a corrected value of the active resonance suppression delay factor; Using the active resonance suppression delay factor correction value to adjust the delay compensation amount of the pulse modulation signal in the grid-connected inverter controller in real time; Delay compensation closed-loop control is performed according to the delay compensation amount, grid-side current time series data after closed-loop control is obtained, and spectrum analysis is performed on the grid-side current time series data to extract high-frequency resonance characteristic values.

7. A method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 6, characterized in that: The high-frequency resonance characteristic values ​​include the resonance frequency point, resonance amplitude and phase mutation interval obtained by dynamic compensation after superimposing the power grid fluctuation.

8. The method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 1, characterized in that: The steps of optimizing the active damping gain parameters with minimizing the resonance peak and maximizing the dynamic response margin as the optimization goals and outputting the optimized set of high-frequency resonance active suppression parameters include: Define the active damping gain parameter as a decision variable and set the optimization range of the active damping gain parameter; With the optimization goals of minimizing the resonance peak and maximizing the dynamic response margin, several particles are randomly generated within the optimization range of the active damping gain parameter to form an initial particle swarm. The active damping gain parameter of each particle is input into the grid-connected inverter controller. The resonance peak is obtained through closed-loop simulation. The phase margin and amplitude margin under the current active damping gain parameter are calculated to obtain the dynamic response margin. Calculating a current fitness value of each particle according to the resonance peak value and the dynamic response margin, and updating an individual extreme value position according to the current fitness value of each particle; Finding a global optimal solution among the individual extreme value positions of all particles, updating the global extreme value position according to the global optimal solution, and updating the speed and position of each particle according to the updated individual extreme value position and the global extreme value position; Repeat the above steps until the fitness value converges, and output the active damping gain parameter combination corresponding to the global optimal extreme value position to form the optimized set of high-frequency resonance active suppression parameters.

9. A method for dynamically suppressing high-frequency resonance of a direct-drive wind turbine grid-connected inverter according to claim 8, characterized in that: The fitness function of each particle is defined as the weighted sum of the inverse of the resonance peak and the dynamic response margin.

10. A high-frequency resonance dynamic suppression system for a direct-drive wind turbine grid-connected inverter, characterized in that: The system comprises: A data acquisition module is used to extract high-frequency resonant natural frequency oscillation distribution data from the collected grid-side current data and grid-connected inverter output voltage waveform data; a damping analysis module, configured to determine an active resonance suppression delay factor based on the high-frequency resonance natural frequency oscillation distribution data, and obtain a preliminary damping value under an equivalent series resistance effect according to the active resonance suppression delay factor and a grid-connected power smoothing constraint condition; a dynamic compensation module, configured to superimpose the preliminary damping value with a pre-acquired grid voltage fluctuation influence factor to obtain a superposition result, and dynamically compensate and adjust the active resonance suppression delay factor according to the superposition result to obtain a high-frequency resonance characteristic value; a drift analysis module, configured to calculate a frequency drift offset based on a difference between the high-frequency resonance characteristic value and a bandwidth upper limit value of the grid-connected inverter controller; a parameter optimization module for optimizing the active damping gain parameters with the optimization objectives of minimizing the resonance peak and maximizing the dynamic response margin when the frequency drift offset exceeds a preset drift threshold range, and outputting an optimized set of high-frequency resonance active suppression parameters; The resonance suppression module is used to perform high-frequency resonance active suppression on a direct-drive wind turbine grid-connected inverter system using the high-frequency resonance active suppression parameter optimization set.

Images