An SVG impedance reconstruction method and system based on adaptive virtual admittance injection

Abstract

The application discloses an SVG impedance reconstruction method and system based on adaptive virtual admittance injection, which adopts an adaptive virtual admittance injection technology, reconstructs SVG impedance, makes it provide positive damping in an oscillation sensitive frequency band, and improves the dynamic stability of a new energy grid-connected system such as an offshore wind power. An adaptive virtual admittance link is connected in parallel in a traditional reactive power control loop of an SVG, the coupling relationship between a port voltage and a current of the SVG is changed, and therefore the reconstruction of the output impedance of the SVG is realized. The core lies in that the parameters of the virtual admittance can be adaptively adjusted according to a main oscillation frequency detected in real time, and the impedance phase of the sensitive frequency band is accurately compensated. On one hand, the introduction of the virtual admittance changes the port characteristics of the SVG, so that the SVG presents the impedance characteristics expected by the design target in a wide frequency band; on the other hand, the frequency adaptive mechanism enables the control method to accurately act on the actual oscillation frequency, so that the suppression effect is better and the robustness is stronger.

Description

Technical Field

[0001] This invention relates to the field of power system stability and control technology, and in particular to an SVG impedance reconstruction method and system based on adaptive virtual admittance injection. Background Technology

[0002] With the integration of high-penetration wind power clusters into the power grid, and especially with the long-distance transmission of large-scale offshore wind power via AC submarine cables, the complex dynamic interaction between the power system and power electronic equipment has induced a new type of broadband oscillation problem.

[0003] Analysis reveals that offshore direct-drive wind farms containing Static Var Generators (SVG) tend to exhibit capacitive negative impedance characteristics in the low-frequency range (e.g., 40-80Hz). When this capacitive negative impedance interacts with the inductive impedance of a weak power grid, a negative damping effect is formed, which is a key factor inducing voltage oscillations. Traditional SVG control strategies primarily focus on reactive power compensation, and the parameter design of their current loop and reactive power outer loop often causes the impedance phase of the equipment to fall into an unstable region in specific frequency bands, exacerbating the risk of system oscillations.

[0004] Therefore, there is an urgent need for a control method that can actively reconstruct the output impedance characteristics of SVG, changing its impedance characteristics in the sensitive frequency band from "capacitive negative damping" to "capacitive positive damping" in order to improve the overall system damping and suppress voltage oscillations. Summary of the Invention

[0005] This invention proposes an SVG impedance reconstruction method based on adaptive virtual admittance injection, which can effectively solve the problems mentioned in the background art.

[0006] To achieve the above objectives, this invention employs the following technical solution: an SVG impedance reconstruction method and system based on adaptive virtual admittance injection. This method reconstructs the output impedance by connecting an adaptive virtual admittance loop in parallel with the traditional reactive power control loop of the SVG, thereby altering the voltage-current coupling relationship at the SVG port. The core of this method lies in the fact that the parameters of the virtual admittance can be adaptively adjusted according to the real-time detected dominant oscillation frequency, accurately compensating for the impedance phase in sensitive frequency bands. On one hand, the introduction of virtual admittance alters the SVG port characteristics, enabling it to exhibit the desired impedance characteristics across a wide frequency range. On the other hand, the frequency adaptive mechanism allows the control method to precisely target the actual oscillation frequency, resulting in better suppression and stronger robustness.

[0007] Furthermore, an SVG impedance reconstruction method based on adaptive virtual admittance injection includes the following steps:

[0008] S1. Connect the grid connection point of the static var generator to the grid voltage source via the grid inductor. The SVG control side samples the grid connection point voltage. The input is sent to the phase-locked loop unit to obtain the grid voltage phase; an adaptive virtual admittance injection branch is connected in parallel between the output terminal of the reactive power outer loop controller and the reference input terminal of the current inner loop. The adaptive virtual admittance injection branch includes an adaptive virtual admittance injection module and a grid connection point voltage acquisition module.

[0009] S2, Sample SVG three-phase output current Together with the phase signal, it is sent to the first coordinate transformation unit to complete the transformation. / Transformation, Output Axis current components and Axis current components This provides the basic signal for subsequent dual closed-loop control and adaptive virtual admittance injection modules.

[0010] S3. The SVG active damping control structure with adaptive virtual admittance control adopts a dual closed-loop control architecture, which includes a reactive power calculation module, a reactive power outer loop control module, and a current inner loop control module. The reactive power calculation module calculates the actual reactive power based on the acquired voltage and current signals. .

[0011] The reactive power reference value is received by the reactive power outer loop controller within the reactive power outer loop control module. Compared with the actual output reactive power of SVG The deviation signal is used to generate the initial value through proportional-integral adjustment. Shaft current reference value .

[0012] S4, the adaptive virtual admittance injection module collects the grid connection point voltage. via virtual admittance An additional current command is generated after the calculation. The additive is used to add to the original Together, they form a new inner current loop reference value; through The axis current reference channel is injected with an adaptive virtual admittance additional instruction, which enables active reshaping of the SVG output impedance, thereby introducing positive damping in the target oscillation frequency band.

[0013] S5, Current Inner Loop Control Module / The error between the shaft current reference value and the actual value is adjusted to generate a modulated voltage signal, which is then sent to the PWM pulse signal generation module after coordinate transformation to generate a switching signal to drive the three-phase full-bridge inverter circuit.

[0014] Advantages compared to existing technologies:

[0015] This invention achieves active reshaping of the SVG output impedance by connecting an adaptive virtual admittance injection module in parallel at the reactive power outer loop output terminal. In the oscillation-sensitive frequency band of 40~80Hz, the original capacitive negative impedance characteristic is transformed into a capacitive positive impedance characteristic, eliminating the negative damping effect at its source and significantly improving the system stability under weak power grid conditions.

[0016] Introducing a frequency-adaptive identification and real-time parameter calculation mechanism enables the virtual admittance gain and phase compensation angle to be adjusted. It can automatically adjust to changes in the dominant oscillation frequency of the system. Compared with fixed parameter compensation, this invention can accurately target the actual oscillation frequency, resulting in better suppression and strong robustness to changes in power grid operation.

[0017] Without this invention, the power oscillation amplitude was as high as 34.27%, and the oscillation decay time was 23.0 ms. After incorporating this invention, the system's dynamic response speed was significantly improved, with the oscillation amplitude decreasing to 29.14% and the decay time shortened to 21.8 ms. See appendix for details. Figure 7 As shown, the active power response is tested under phase perturbation.

[0018] Without this invention, the voltage oscillation amplitude was 19.89% and the decay time was 0.4s; after incorporating this invention, the oscillation amplitude decreased to 11.86%, the decay time was shortened to 0.2s, and the system oscillation converged faster. See appendix for details. Figure 8 As shown, the voltage response at the grid connection point is tested under voltage disturbance.

[0019] After implementing this invention, the voltage oscillation of the system under wind speed fluctuation conditions was significantly suppressed, and the response waveform stabilized extremely quickly, proving that this invention has an excellent suppression effect on broadband oscillations caused by fluctuations in new energy output. See appendix for details. Figure 9 As shown, the voltage response at the grid connection point was tested under wind speed disturbance.

[0020] The system damping ratio increased from 0.3026 to 0.5684 under grid phase disturbances, an increase of approximately 87.8%; under voltage disturbances, the damping ratio increased from 0.3575 to 0.7849, an increase of approximately 119.5%; and under wind speed disturbances from the sub-wind farm, the damping ratio increased from 0.2283 to 0.4617, an increase of approximately 102.2%. See appendix for details. Figure 10 The system damping ratios under different disturbance modes are summarized. Attached Figure Description

[0021] Figure 1 Flowchart for injecting adaptive virtual admittance;

[0022] Figure 2 This is a topology diagram of the SVG structure.

[0023] Figure 3 A schematic diagram of adaptive active damping access in the outer loop of an SVG system with adaptive virtual admittance control;

[0024] Figure 4 is a block diagram of SVG adaptive virtual admittance control;

[0025] Figure 5 is a schematic diagram of the adaptive virtual admittance injection module;

[0026] Figure 6 shows a comparison of the impedance characteristics of SVG under different control parameters;

[0027] Figure 7 is a comparison of the active power response of the system with and without the present invention under phase disturbance;

[0028] Figure 8 is a comparison of the grid connection voltage response of the system with and without the present invention under voltage disturbance.

[0029] Figure 9 is a comparison of the grid connection voltage response of the system with and without the present invention under wind speed disturbance;

[0030] Figure 10 shows a comparative analysis of the system damping ratio under different disturbance modes.

[0031] 1. Three-phase full-bridge inverter circuit; 2. DC-side capacitor; 3. Filter inductor; 4. Grid inductor; 5. AC-side capacitor; 6. Damping resistor; 7. Phase-locked loop unit; 8. First coordinate transformation unit; 201. Adaptive virtual admittance injection branch; 202. Power outer loop controller; 203. Current inner loop controller; 205. Shaft decoupling compensation module, 206. 207. Axis decoupling compensation module; 208. Second coordinate transformation unit; 309. PWM pulse generation module; 300. Reactive power calculation module; 301. Reactive power outer loop control module; 302. Current inner loop control module; 303. PWM pulse signal generation module; 400. Grid connection point voltage acquisition module; 401. Adaptive virtual admittance injection module. Detailed Implementation

[0032] Example 1, refer to Appendix Figures 1-10 To make the technical solution of the present invention clearer, further explanation is provided below in conjunction with the accompanying drawings and specific examples. It should be understood that the embodiments listed herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0033] In the following comparative experiments, "without implementing this invention" refers to using the traditional dual closed-loop control strategy without a virtual admittance injection module; "implementing this invention" refers to using the adaptive virtual admittance injection control strategy proposed in this patent.

[0034] like Figure 1The diagram shown is a flowchart of the adaptive virtual admittance injection process. First, the grid connection point voltage acquisition module acquires the voltage signal at the SVG grid connection point in real time. The voltage component reflecting the oscillation or disturbance characteristics of the system is extracted and used as the input signal for subsequent virtual admittance calculation.

[0035] Secondly, frequency identification is performed on the grid connection point voltage disturbance signal to obtain the dominant oscillation frequency. This is used to determine the oscillation frequency bands that need to be suppressed in the current system. Then, it is determined whether impedance reconstruction is needed. When the identified dominant oscillation frequency falls into the preset target frequency band, or the voltage disturbance amplitude, impedance amplitude, or oscillation risk index at the corresponding frequency exceeds a preset threshold, it is determined that impedance reconstruction is needed; otherwise, the original control parameters remain unchanged, and real-time monitoring continues. When impedance reconstruction is determined to be needed, virtual admittance parameters are calculated based on the dominant oscillation frequency, with the goal of making the equivalent output impedance of the SVG at the target oscillation frequency exhibit impedance characteristics conducive to system stability. Subsequently, the virtual admittance model is updated based on the calculated virtual admittance parameters, so that the virtual admittance model has the desired amplitude-frequency and phase-frequency characteristics in the target oscillation frequency band. Through real-time updates, the control system can adaptively adjust the impedance reconstruction effect according to changes in the oscillation frequency. Afterwards, based on the updated virtual admittance model and the grid connection point voltage disturbance signal, a... Shaft-attached current reference command.

[0036] Finally, the q-axis additional current reference command is superimposed on the q-axis current reference value output by the reactive power outer loop. The compensated After the shaft current reference value is input to the inner loop control module, the equivalent relationship between the SVG port voltage and the output current is changed, thereby realizing the adaptive reconstruction of the SVG output impedance, introducing positive damping in the target oscillation frequency band, and suppressing subsynchronous or supersynchronous voltage oscillation.

[0037] Figure 2 shows the topology of the SVG main circuit and sampling control signal flow, which includes adaptive virtual admittance control. The Static Var Generator (SVG) main circuit consists of a three-phase full-bridge inverter circuit 1, a DC-side capacitor 2, a filter inductor 3, an AC-side capacitor 5, and a damping resistor 6; the grid connection point is connected to the grid voltage source V via the grid inductor 4. S The SVG control side samples the grid connection point voltage. The input is sent to the phase-locked loop unit to obtain the grid voltage phase; simultaneously, the three-phase output current of the SVG is sampled. Together with the phase signal, it is sent to the first coordinate transformation unit 8 to complete the transformation. / Transformation, Output Axis current components and Axis current components This provides the basic signal for subsequent dual closed-loop control and adaptive virtual admittance injection modules.

[0038] The feature that distinguishes this invention from the prior art is that an adaptive virtual admittance injection module 401 is connected in parallel between the output terminal of the reactive power outer loop controller and the reference input terminal of the current inner loop. This module collects the voltage at the grid connection point. via virtual admittance An additional current command is generated after the calculation. The additive is used to add to the original The above together constitute a new reference value for the inner current loop, as shown in equation (1):

[0039] = (1)

[0040] In this way, the voltage-current coupling relationship at the SVG port is changed, and the output impedance is reconstructed. Essentially, it is an impedance reshaping control strategy based on virtual admittance injection.

[0041] Figure 3 shows a schematic diagram of the SVG outer loop active damping control structure with adaptive virtual admittance control.

[0042] The control structure adopts a dual closed-loop control architecture, in which the reactive power outer loop controller 202 receives the reactive power reference value. Compared with the actual output reactive power of SVG The deviation signal is used to generate the initial value through proportional-integral adjustment. Shaft current reference value In the inner current control channel, Shaft current reference value and actual current The shaft current reference value and the actual current are respectively sent to the corresponding current inner loop controller 203 for adjustment through a comparison circuit.

[0043] The core improvement of this invention lies in the fact that an additional current command signal generated by the adaptive virtual admittance injection branch 201 is connected in parallel between the reactive power outer loop output terminal and the current inner loop reference input terminal, and is superimposed onto the initial current through an adder stage. This forms a new inner current loop reference value. Shaft decoupling compensation module 205 The shaft decoupling compensation module 206 respectively implements the following: , The feedback correction is performed, and the corrected signal is converted into a modulation signal by the second coordinate transformation unit 207, and finally sent to the PWM pulse generation module 208 to generate a switching signal to drive the three-phase full-bridge inverter circuit.

[0044] This structure is achieved through... The axis current reference channel is injected with an adaptive virtual admittance additional instruction, which realizes the active reshaping of the SVG output impedance, thereby introducing positive damping in the target oscillation frequency band and suppressing subsynchronous / supersynchronous voltage oscillation.

[0045] As shown in Figure 4, the SVG control adopts a dual closed-loop structure. The reactive power calculation module 301 calculates the actual reactive power based on the acquired voltage and current signals. The reactive power outer loop control module 302 will use the reactive power reference value. With actual reactive power The difference is processed, and an initial value is generated through a proportional-integral controller. Shaft current reference value Current inner loop control module 303 / Reference value and actual value of shaft current ( / The error is adjusted to generate a modulated voltage signal, which is then sent to the PWM pulse signal generation module 304 after coordinate transformation to generate a switching signal to drive the three-phase full-bridge inverter circuit.

[0046] The distinguishing feature of this invention from existing technologies is that an adaptive virtual admittance injection module 401 and a grid connection point voltage acquisition module 400 are connected in parallel between the output terminal of the reactive power outer loop controller and the reference input terminal of the current inner loop controller. The grid connection point voltage acquisition module outputs... Through the adaptive virtual admittance injection module, via virtual admittance An additional current command is generated after the calculation. The compensated total reference current is generated by the adder. In this way, the voltage-current coupling relationship at the SVG port is changed, thereby achieving the reconstruction of the output impedance.

[0047] Figure 5 This is a schematic diagram of the adaptive virtual admittance injection module. The module mainly includes: a filter bank (composed of a low-pass filter 411, a band-pass filter 412, and a high-pass filter 413), a frequency identification module 420, a parameter calculation module 430, and a virtual admittance execution module 440.

[0048] The filter bank (low-pass filter 411, band-pass filter 412, and high-pass filter 413) is used to separate the acquired grid connection point voltage. Different frequency band disturbance components in the.

[0049] The frequency identification module 420 extracts and identifies the frequency of the dominant oscillation mode with the largest amplitude in the system in real time through a phase-locked loop or a fast Fourier transform (FFT) algorithm. .

[0050] Parameter calculation module 430 calculates based on frequency Real-time calculation of virtual admittance Gain coefficient and phase compensation angle Its design goal is to enable high frequency... At this point, the equivalent output impedance of the SVG meets the pre-set stability criterion (i.e., the phase difference with the mains impedance is less than 180°).

[0051] The virtual admittance execution module 440 generates a specific virtual admittance transfer function model based on the real-time updated parameters, and its output is processed by Δi q After the output module, the additional current command is obtained. This is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port.

[0052] The virtual admittance transfer function model :

[0053] (2)

[0054] in, According to frequency The gain is adaptively adjusted; T1 and T2 are the time constants of the phase compensation stage, and their ratio determines the magnitude and frequency band of the phase compensation. They can be configured as multiple stages in series according to actual needs. The value range of T1 / T2 is 0.1~10.

[0055] Additional current command output by virtual admittance execution module 440 After being added by the adder and the original After being superimposed, the current is input to the inner current loop control module 303. The injection of this additional current command is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port, thereby reshaping the SVG output impedance.

[0056] To reveal the reconstruction mechanism of the SVG output impedance in this invention, a small-signal sequence impedance model of the SVG needs to be established. In the dq rotating coordinate system, the dynamic equation of the SVG's main circuit can be expressed as:

[0057] (3)

[0058] In the formula, For filtering inductors, Equivalent resistance The fundamental angular frequency, , For the grid connection point voltage / Axial components. , The inverter output voltage / Axial components.

[0059] In traditional SVG, the control strategy is based on a dual closed-loop structure consisting of a voltage outer loop and a current inner loop. The current inner loop is implemented using a PI controller, and the corresponding transfer function is:

[0060] (4)

[0061] The reactive power outer loop also uses PI control:

[0062] (5)

[0063] Without considering virtual admittance injection, the output admittance of SVG can be derived as follows:

[0064] (6)

[0065] in This is the equivalent transfer function of the inverter. Studies have shown that when system parameters are not set correctly, in the 40~80Hz frequency band, It exhibits capacitive characteristics, and the phase exceeds the range of [-90°, 90°], that is, it exhibits capacitive negative damping characteristics.

[0066] This invention connects a virtual admittance injection module in parallel at the output terminal of the reactive power outer loop to generate an additional current command. The relationship between this additional instruction and the grid connection point voltage can be expressed as:

[0067] (7)

[0068] The corresponding output impedance is:

[0069] (8)

[0070] Through design The amplitude and phase frequency characteristics can make It exhibits the desired impedance characteristics in the target oscillation frequency band.

[0071] The parameter adaptive adjustment strategy of this invention introduces a frequency adaptive mechanism to achieve precise suppression of the actual oscillation frequency of the system. Let the dominant oscillation frequency detected by the frequency identification unit be... The corresponding angular frequency is The parameter adaptive calculation unit is based on... Adjust the virtual admittance parameters.

[0072] Gain Adaptive:

[0073] (9)

[0074] Phase compensation adaptive:

[0075] In frequency Provide the required phase compensation angle at the location The time constants T1 and T2 must satisfy:

[0076] (10)

[0077] With the T1 / T2 ratio fixed, adjust the absolute value of the time constant according to the target frequency:

[0078] (11)

[0079] Stability criterion after adaptive impedance reconstruction: After introducing adaptive virtual admittance, the stability of the closed-loop system composed of SVG and the power grid is determined by the following Nyquist criterion:

[0080] (12)

[0081] It satisfies the Nyquist stability criterion.

[0082] in This is the equivalent impedance corresponding to the virtual admittance. (The rest of the text appears to be a typo and can be omitted.) Envoy:

[0083] (13)

[0084] This ensures that the phase margin of the system is positive, thereby achieving positive damping injection and effectively suppressing voltage oscillations.

[0085] As shown in Figure 6, the amplitude-frequency and phase-frequency characteristic curves of the SVG impedance are plotted under different current loop and reactive power outer loop control parameters. The comparison shows that traditional control parameter settings cause the SVG to exhibit capacitive negative impedance characteristics with a phase exceeding the range of [-90°, 90°] in the 40-80Hz frequency band. After introducing the strategy of this invention, the impedance characteristics are reshaped in the sensitive frequency band.

[0086] Example 2: An SVG impedance reconstruction system based on adaptive virtual admittance injection, including a static var generator and an SVG control system.

[0087] The SVG control system includes a reactive power outer loop controller and a current inner loop controller.

[0088] The main circuit of the Static Var Generator (SVG) consists of a three-phase full-bridge inverter circuit 1, a DC-side capacitor 2, a filter inductor 3, an AC-side capacitor 5, and a damping resistor 6.

[0089] The grid connection point of the static var generator is connected to the grid voltage source via grid inductor 4. The SVG control side samples the grid connection point voltage. The input is sent to phase-locked unit 7 to obtain the grid voltage phase;

[0090] An additional current command signal generated by the adaptive virtual admittance injection branch 201 is connected in parallel between the output terminal of the reactive power outer loop controller and the reference input terminal of the current inner loop.

[0091] The adaptive virtual admittance injection branch 201 includes an adaptive virtual admittance injection module 401 and a grid connection point voltage acquisition module 400. The grid connection point voltage acquisition module outputs... Through the adaptive virtual admittance injection module, via virtual admittance An additional current command is generated after the calculation. The compensated total reference current is generated by the adder. .

[0092] An additional current command signal generated by the adaptive virtual admittance injection branch 201 is connected in parallel between the reactive power outer loop output terminal and the current inner loop reference input terminal.

[0093] The SVG control system adopts a dual closed-loop structure, including a reactive power calculation module 301, a reactive power outer loop control module 302, and a current inner loop control module 303.

[0094] The reactive power calculation module 301 calculates the actual reactive power based on the collected voltage and current signals. .

[0095] The reactive power outer loop control module 302 will use the reactive power reference value With actual reactive power The difference is processed, and an initial value is generated through a proportional-integral controller. Shaft current reference value Current inner loop control module 303 / Reference value and actual value of shaft current ( / The error is adjusted to generate a modulated voltage signal, which is then sent to the PWM pulse signal generation module 304 after coordinate transformation to generate a switching signal to drive the three-phase full-bridge inverter circuit.

[0096] The adaptive virtual admittance injection module includes a filter bank, a frequency identification module 420, a parameter calculation module 430, and a virtual admittance execution module 440.

[0097] Among them, the filter bank is used to separate the acquired grid connection point voltage. Different frequency band disturbance components in the.

[0098] The filter bank consists of a low-pass filter 411, a band-pass filter 412, and a high-pass filter 413.

[0099] The frequency identification module 420 extracts and identifies the frequency of the dominant oscillation mode with the largest amplitude in the system in real time through a phase-locked loop or a fast Fourier transform (FFT) algorithm. .

[0100] Parameter calculation module 430 calculates based on frequency Real-time calculation of virtual admittance Gain coefficient and phase compensation angle Its design goal is to enable high frequency... At this point, the equivalent output impedance of the SVG meets the pre-set stability criterion (i.e., the phase difference with the mains impedance is less than 180°).

[0101] The virtual admittance execution module 440 generates a specific virtual admittance transfer function model based on real-time updated parameters, and its output is processed... After the output module, the additional current command is obtained. This is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port.

[0102] Example 3: An SVG impedance reconstruction method based on adaptive virtual admittance injection, comprising the following steps:

[0103] S1. Connect the grid connection point of the static var generator to the grid voltage source via grid inductor 4. The SVG control side samples the grid connection point voltage. The input is sent to the phase-locked unit 7 to obtain the grid voltage phase. An adaptive virtual admittance injection branch 201 is connected in parallel between the output terminal of the reactive power outer loop controller and the reference input terminal of the current inner loop. The adaptive virtual admittance injection branch 201 includes an adaptive virtual admittance injection module 401 and a grid connection point voltage acquisition module 400.

[0104] S2, Sample SVG three-phase output current Together with the phase signal, it is sent to the first coordinate transformation unit 8 to complete the transformation. / Transformation, Output Axis current components and Axis current components This provides the basic signal for the subsequent dual closed-loop control and adaptive virtual admittance injection module 401.

[0105] S3. The SVG active damping control structure with adaptive virtual admittance control adopts a dual closed-loop control architecture, which includes a reactive power calculation module 301, a reactive power outer loop control module 302, and a current inner loop control module 303. The reactive power calculation module 301 calculates the actual reactive power based on the acquired voltage and current signals. .

[0106] The reactive power reference value is received by the reactive power outer loop controller 202 within the reactive power outer loop control module 302. Compared with the actual output reactive power of SVG The deviation signal is used to generate the initial value through proportional-integral adjustment. Shaft current reference value .

[0107] S4, the adaptive virtual admittance injection module 401 collects the grid connection point voltage. via virtual admittance An additional current command is generated after the calculation. The additive is used to add to the original Together, they form a new inner current loop reference value. Through... The adaptive virtual admittance additional instruction injected into the shaft current reference channel realizes the active reshaping of the SVG output impedance, thereby introducing positive damping in the target oscillation frequency band and suppressing subsynchronous or supersynchronous voltage oscillation.

[0108] The new reference value for the inner current loop is shown in the following formula:

[0109] = (1)

[0110] In this way, the voltage-current coupling relationship at the SVG port is changed, and the output impedance is reconstructed. Essentially, it is an impedance reshaping control strategy based on virtual admittance injection.

[0111] The adaptive virtual admittance injection module 401 includes a filter bank, a frequency identification module 420, a parameter calculation module 430, and a virtual admittance execution module 440.

[0112] Among them, the filter bank is used to separate the acquired grid connection point voltage. Different frequency band disturbance components in the.

[0113] The filter bank consists of a low-pass filter 411, a band-pass filter 412, and a high-pass filter 413.

[0114] The frequency identification module 420 extracts and identifies the frequency of the dominant oscillation mode with the largest amplitude in the system in real time through a phase-locked loop or a fast Fourier transform (FFT) algorithm. .

[0115] Parameter calculation module 430 calculates based on frequency Real-time calculation of virtual admittance Gain coefficient and phase compensation angle Its design goal is to enable high frequency performance. At this point, the equivalent output impedance of the SVG satisfies the pre-set stability criterion.

[0116] The virtual admittance execution module 440 generates a specific virtual admittance transfer function model based on real-time updated parameters, and its output is processed... After the output module, the additional current command is obtained. This is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port.

[0117] The virtual admittance transfer function model :

[0118] (2)

[0119] in, According to frequency The gain is adaptively adjusted. T1 and T2 are the time constants of the phase compensation stage. Their ratio determines the magnitude and bandwidth of the phase compensation. They can be configured as multiple stages in series according to actual needs. The value range of T1 / T2 is usually 0.1~10.

[0120] Additional current command output by virtual admittance execution module 440 After being added by the adder and the original After being superimposed, the current is input to the inner current loop control module 303. The injection of this additional current command is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port, thereby reshaping the SVG output impedance.

[0121] The reshaping of the SVG output impedance includes the following steps:

[0122] Establish the small-signal sequence impedance model of the SVG. In the dq rotating coordinate system, the dynamic equation of the SVG's main circuit can be expressed as:

[0123] (3)

[0124] In the formula, For filtering inductors, Where ω is the equivalent resistance and ω is the fundamental angular frequency. , For the grid connection point voltage / Axial components. , The inverter output voltage / Axial components.

[0125] In traditional SVG, the control strategy is based on a dual closed-loop structure consisting of a voltage outer loop and a current inner loop. The current inner loop is implemented using a PI controller, and the corresponding transfer function is:

[0126] (4)

[0127] The reactive power outer loop also uses PI control:

[0128] (5)

[0129] Without considering virtual admittance injection, the output admittance of SVG can be derived as follows:

[0130] (6)

[0131] in This is the equivalent transfer function of the inverter. Studies have shown that when system parameters are not set correctly, in the 40~80Hz frequency band, It exhibits capacitive characteristics, and the phase exceeds the range of [-90°, 90°], that is, it exhibits capacitive negative damping characteristics.

[0132] A virtual admittance injection module is connected in parallel at the output of the reactive power outer loop to generate an additional current command. The relationship between this additional instruction and the grid connection point voltage can be expressed as:

[0133] (7)

[0134] The corresponding output impedance is:

[0135] (8)

[0136] Through design The amplitude and phase frequency characteristics can make It exhibits the desired impedance characteristics in the target oscillation frequency band.

[0137] To achieve precise suppression of the actual oscillation frequency of the system, a frequency adaptive adjustment strategy is introduced. Let the dominant oscillation frequency detected by the frequency identification unit be... The corresponding angular frequency is The parameter adaptive calculation unit is based on Adjust the virtual admittance parameters.

[0138] Gain Adaptive:

[0139] (9)

[0140] Phase compensation adaptive:

[0141] In frequency Provide the required phase compensation angle at the location The time constants T1 and T2 must satisfy:

[0142] (10)

[0143] With the T1 / T2 ratio fixed, adjust the absolute value of the time constant according to the target frequency:

[0144] (11)

[0145] Stability criterion after adaptive impedance reconstruction: After introducing adaptive virtual admittance, the stability of the closed-loop system composed of SVG and the power grid is determined by the following Nyquist criterion:

[0146] (12)

[0147] It satisfies the Nyquist stability criterion.

[0148] in This is the equivalent impedance corresponding to the virtual admittance. (The rest of the text appears to be a typo and can be omitted.) Envoy:

[0149] (13)

[0150] This ensures that the phase margin of the system is positive, thereby achieving positive damping injection and effectively suppressing voltage oscillations.

[0151] S5, Current Inner Loop Control Module 303 / Reference value and actual value of shaft current ( / The error is adjusted to generate a modulated voltage signal, which is then sent to the PWM pulse signal generation module 304 after coordinate transformation to generate a switching signal to drive the three-phase full-bridge inverter circuit.

[0152] The current inner loop control module 303 includes a current inner loop controller 203, Shaft decoupling compensation module 205 Shaft decoupling compensation module 206.

[0153] The PWM pulse signal generation module 304 includes a second coordinate transformation unit 207 and a PWM pulse generation module 208.

[0154] S5.1, In the current inner loop control channel, Shaft current reference value and actual current The shaft current reference value and the actual current are respectively sent to the corresponding current inner loop controller 203 for adjustment through a comparison circuit. Shaft decoupling compensation module 205 The shaft decoupling compensation module 206 respectively implements the following: , Feedback correction.

[0155] S5.2 The corrected signal is converted into a modulation signal by the second coordinate transformation unit 207 and finally sent to the PWM pulse generation module 208 to generate a switching signal to drive the three-phase full-bridge inverter circuit.

Claims

1. An SVG impedance reconstruction method based on adaptive virtual admittance injection, characterized in that, Includes the following steps: S1. The grid connection point of the static var generator is connected to the grid voltage source VS through the grid inductor (4). The SVG control side samples the grid connection point voltage. Input to the phase-locked unit (7) to obtain the grid voltage phase; between the output terminal of the reactive power outer loop controller and the reference input terminal of the current inner loop, an adaptive virtual admittance injection branch (201) is connected in parallel. The adaptive virtual admittance injection branch (201) includes an adaptive virtual admittance injection module (401) and a grid connection point voltage acquisition module (400). S2, Sample SVG three-phase output current Together with the phase signal, it is sent to the first coordinate transformation unit (8) to complete the transformation. / Transformation, Output Axis current components and Axis current components This provides the basic signal for the subsequent dual closed-loop control and adaptive virtual admittance injection module (401); S3. The SVG active damping control structure with adaptive virtual admittance control adopts a dual closed-loop control architecture, which includes a reactive power calculation module (301), a reactive power outer loop control module (302), and a current inner loop control module (303). The reactive power calculation module (301) calculates the actual reactive power based on the collected voltage and current signals. ; The reactive power reference value is received by the reactive power outer loop controller (202) within the reactive power outer loop control module (302). Compared with the actual output reactive power of SVG The deviation signal is used to generate the initial value through proportional-integral adjustment. Shaft current reference value ; S4, Adaptive Virtual Admittance Injection Module (401) collects grid connection point voltage. via virtual admittance An additional current command is generated after the calculation. The additive is used to add to the original Together, they form a new inner current loop reference value; through The adaptive virtual admittance additional instruction injected into the shaft current reference channel realizes the active reshaping of the SVG output impedance, thereby introducing positive damping in the target oscillation frequency band; S5, Current Inner Loop Control Module (303) / The error between the shaft current reference value and the actual value is adjusted to generate a modulation voltage signal, which is then sent to the PWM pulse signal generation module (304) after coordinate transformation to generate a switching signal to drive the three-phase full-bridge inverter circuit.

2. The SVG impedance reconstruction method based on adaptive virtual admittance injection according to claim 1, characterized in that: S5 includes the following steps: S5.1, In the current inner loop control channel, Shaft current reference value and actual current The shaft current reference value and the actual current are respectively sent to the corresponding current inner loop controller (203) for adjustment through the comparison link; Shaft decoupling compensation module (205) The shaft decoupling compensation module (206) respectively implements the following: , Feedback correction; S5.2 The corrected signal is converted into a modulation signal by the second coordinate transformation unit (207) and finally sent to the PWM pulse generation module (208) to generate a switching signal to drive the three-phase full-bridge inverter circuit.

3. The SVG impedance reconstruction method based on adaptive virtual admittance injection according to claim 1, characterized in that: The S4 includes the following: the adaptive virtual admittance injection module (401) includes a filter bank, a frequency identification module (420), a parameter calculation module (430), and a virtual admittance execution module (440). Among them, the filter bank is used to separate the acquired grid connection point voltage. Different frequency band disturbance components; The filter bank consists of a low-pass filter (411), a band-pass filter (412), and a high-pass filter (413); The frequency identification module (420) extracts and identifies the frequency of the dominant oscillation mode with the largest amplitude in the system in real time through a phase-locked loop or a fast Fourier transform (FFT) algorithm. ; The parameter calculation module (430) calculates the frequency based on the frequency. Real-time calculation of virtual admittance Gain coefficient and phase compensation angle Its design goal is to enable high frequency At this point, the equivalent output impedance of the SVG satisfies the preset stability criterion; The virtual admittance execution module (440) generates a specific virtual admittance transfer function model based on the real-time updated parameters, and its output is processed... After the output module, the additional current command is obtained. This is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port; The virtual admittance transfer function model : (2) in, According to frequency The gain is adaptively adjusted; T1 and T2 are the time constants of the phase compensation stage, and their ratio determines the magnitude and bandwidth of the phase compensation. They can be configured as multiple stages in series according to actual needs. The value range of T1 / T2 is 0.1~10. Additional current command output by the virtual admittance execution module (440) After being added by the adder and the original After superposition, the input is sent to the inner current loop control module (303); the injection of this additional current command is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port, thereby realizing the reshaping of the SVG output impedance.

4. The SVG impedance reconstruction method based on adaptive virtual admittance injection according to claim 1, characterized in that: The new reference value for the inner current loop in S4 is shown in the following formula: = (1)。 5. The SVG impedance reconstruction method based on adaptive virtual admittance injection according to claim 3, characterized in that: The reshaping of the SVG output impedance includes the following steps: Establish the small-signal sequence impedance model of the SVG; in the dq rotating coordinate system, the dynamic equation of the main circuit of the SVG can be expressed as: (3) In the formula, For filtering inductors, Where ω is the equivalent resistance and ω is the fundamental angular frequency. , For the grid connection point voltage / Axial components; , For the inverter output voltage / Axial components; In traditional SVG, the control strategy is based on a dual closed-loop structure consisting of a voltage outer loop and a current inner loop; the current inner loop is implemented using a PI controller, and the corresponding transfer function is: (4) The reactive power outer loop also uses PI control: (5) Without considering virtual admittance injection, the output admittance of SVG can be derived as follows: (6) in This is the equivalent transfer function of the inverter; research shows that when system parameters are not set correctly, in the 40~80Hz frequency band, It exhibits capacitive characteristics, and the phase exceeds the range of [-90°, 90°], that is, it exhibits capacitive negative damping characteristics; A virtual admittance injection module is connected in parallel at the output of the reactive power outer loop controller to generate an additional current command. The relationship between this additional instruction and the grid connection point voltage can be expressed as follows: (7) The corresponding output impedance is: (8) Through design The amplitude and phase frequency characteristics can make It exhibits the desired impedance characteristics in the target oscillation frequency band; To achieve precise suppression of the actual oscillation frequency of the system, a frequency adaptive mechanism is introduced as a parameter adaptive adjustment strategy; let... The dominant oscillation frequency detected by the frequency identification unit is The corresponding angular frequency is The parameter adaptive calculation unit is based on Adjust the virtual admittance parameters; Gain Adaptive: (9) Phase compensation adaptive: In frequency Provide the required phase compensation angle at the location The time constants T1 and T2 must satisfy: (10) With the T1 / T2 ratio fixed, adjust the absolute value of the time constant according to the target frequency: (11) Stability criterion after adaptive impedance reconstruction: After introducing adaptive virtual admittance, the stability of the closed-loop system composed of SVG and the power grid is determined by the following Nyquist criterion: (12) Satisfies the Nyquist stability criterion; in The equivalent impedance corresponding to the virtual admittance; by the oscillation frequency Envoy: (13) This ensures that the phase margin of the system is positive, thereby achieving positive damping injection and effectively suppressing voltage oscillations.

6. The SVG impedance reconstruction method based on adaptive virtual admittance injection according to claim 1, characterized in that: Including static var generators and SVG control systems; The SVG control system includes a reactive power outer loop controller and a current inner loop controller; The main circuit of the Static Var Generator (SVG) consists of a three-phase full-bridge inverter circuit (1), a DC side capacitor (2), a filter inductor (3), an AC side capacitor (5), and a damping resistor (6). The grid connection point of the static var generator is connected to the grid voltage source via the grid inductor (4). The SVG control side samples the grid connection point voltage. The input is sent to the phase-locked unit (7) to obtain the grid voltage phase; An adaptive virtual admittance injection module (401) and a grid connection point voltage acquisition module (400) are connected in parallel between the output of the reactive power outer loop controller and the reference input of the current inner loop; the output of the grid connection point voltage acquisition module is... Through the adaptive virtual admittance injection module (401), via virtual admittance An additional current command is generated after the calculation. The compensated total reference current is generated by the adder. ; An additional current command signal generated by the adaptive virtual admittance injection branch (201) is connected in parallel between the output terminal of the reactive power outer loop controller and the reference input terminal of the current inner loop. The SVG control system adopts a dual closed-loop structure, including a reactive power calculation module (301), a reactive power outer loop control module (302), and a current inner loop control module (303). The reactive power calculation module (301) calculates the actual reactive power based on the collected voltage and current signals. ; The reactive power outer loop control module (302) will use the reactive power reference value With actual reactive power The difference is processed, and an initial value is generated through a proportional-integral controller. Shaft current reference value The current inner loop control module (303) controls the current. / The error between the shaft current reference value and the actual value is adjusted to generate a modulation voltage signal, which is then sent to the PWM pulse signal generation module (304) after coordinate transformation to generate a switching signal to drive the three-phase full-bridge inverter circuit. The adaptive virtual admittance injection module (401) includes a filter bank, a frequency identification module (420), a parameter calculation module (430), and a virtual admittance execution module (440); Among them, the filter bank is used to separate the acquired grid connection point voltage. Different frequency band disturbance components; The filter bank consists of a low-pass filter (411), a band-pass filter (412), and a high-pass filter (413); The frequency identification module (420) extracts and identifies the frequency of the dominant oscillation mode with the largest amplitude in the system in real time through a phase-locked loop or a fast Fourier transform (FFT) algorithm. ; The parameter calculation module (430) calculates the frequency based on the frequency. Real-time calculation of virtual admittance Gain coefficient and phase compensation angle Its design goal is to enable high frequency At this point, the equivalent output impedance of the SVG satisfies the preset stability criterion; The virtual admittance execution module (440) generates a specific virtual admittance transfer function model based on the real-time updated parameters, and its output is processed... After the output module, the additional current command is obtained. This is equivalent to connecting an adaptively adjustable virtual admittance in parallel at the SVG port.

Images