Virtual synchronous machine control method based on feedback of active fractional order high-pass filter

Abstract

The invention discloses a virtual synchronous machine control method based on active fractional order high-pass filter feedback, and the method comprises the steps: introducing a power dynamic compensation link based on the active fractional order high-pass filter feedback into a conventional virtual synchronous machine grid-connected control structure; according to the method, the dynamic response performance of the active power and the output frequency of the virtual synchronous machine grid-connected system is improved, and the parameters of a fractional order high-pass filter and a feedback compensation coefficient are adjusted; the method can effectively suppress dynamic oscillation and overshoot of active power and output frequency of the virtual synchronous machine grid-connected system caused by active reference instruction change or power grid frequency disturbance, and has the advantages of higher control degree of freedom, more flexible parameter adjustment and smaller output frequency response overshoot amplitude compared with a traditional virtual synchronous machine grid-connected control method. The method can be applied to the field of virtual synchronous machine grid-connected control in the power electronic technology.

Description

Technical Field

[0001] This invention relates to the field of virtual synchronous machine control technology, and in particular to a virtual synchronous machine control method based on active power fractional-order high-pass filter feedback. It is applicable to the field of virtual synchronous machine grid-connected control in power electronics technology and the field of microgrid integration into distribution networks containing such virtual synchronous machines. Background Technology

[0002] With the rapid growth of global renewable energy grid connection, the proportion of traditional synchronous generator power sources is declining, and the dynamic characteristics of power systems are changing significantly, with the problems of "low immunity and weak support" becoming increasingly prominent. Against this backdrop, Virtual Synchronous Generators (VSGs) can provide voltage and inertia support; however, the active power and output frequency of VSG grid-connected systems are prone to dynamic oscillations and overshoot under active power reference commands or grid frequency disturbances. Furthermore, introducing a dynamic power compensation stage based on the differential feedback of an active power high-pass filter into the VSG grid-connected control system, and optimizing the feedback compensation system, can effectively suppress the dynamic oscillations and overshoot of VSG grid-connected active power and output frequency. However, the ability to suppress VSG grid-connected active power overshoot and output frequency response overshoot still needs further improvement.

[0003] To address this, various studies have been conducted, such as the article entitled "Suppression Strategy for Transient Power Oscillation of Virtual Synchronous Generator Considering Overshoot," published in Automation of Electric Power Systems, Vol. 46, No. 11, 2022, pp. 131-141. This article proposes a method for suppressing the dynamic oscillation of VSG grid-connected active power based on the differential feedback of an active power high-pass filter. This method can effectively enhance the transient damping of the closed-loop control of VSG grid-connected active power and reduce the dynamic oscillation and power overshoot of VSG grid-connected active power. However, the differential feedback of the active power high-pass filter introduced in this method is still limited in terms of improving the degree of freedom of control and the flexibility of parameter adjustment, and the suppression effect on the overshoot of VSG grid-connected active power and the overshoot of the output frequency response still needs further optimization.

[0004] The article, titled “Fractional-order virtual inertia control and parameter tuning for energy-storage system in low-inertia power grid”, ZENG YK, YANG QF, LIN YJ, et al., Protection and Control of Modern Power Systems, 2024, 9(5), 70-83 (“Fractional-order virtual inertia control and parameter tuning for energy-storage system in low-inertia power grid”, Protection and Control of Modern Power Systems, 2024, Vol. 9, No. 5, pp. 70-83), proposes a VSG control and parameter tuning method based on fractional-order virtual inertia, which can improve the control degree of freedom of VSG grid-connected system and optimize the dynamic response performance of VSG grid-connected active power and output frequency by determining the stability domain of VSG control parameters through stable boundary trajectory. However, it is still difficult to balance the active power dynamic response performance and inertia support capability of its grid-connected system, and the overshoot suppression performance of output frequency response still needs to be further improved.

[0005] The article, titled "Control Strategy for Fractional-Order Virtual Synchronous Machine Based on RBF Neural Network," published in the *Journal of Electric Power Systems and Automation*, Vol. 37, No. 9, 2025, pp. 101-108, establishes a mathematical model of a fractional-order virtual synchronous machine. By introducing adjustable fractional-order parameters, it increases the control degrees of freedom and parameter adjustment flexibility of the VSG grid-connected system. Furthermore, it designs a radial basis function neural network to adaptively adjust the virtual inertia coefficient and virtual damping coefficient of the VSG online, effectively suppressing the oscillation and overshoot of the VSG's active power and output frequency. However, it suffers from drawbacks such as nonlinear parameter changes, difficulty in parameter tuning, and large overshoot amplitude in the output frequency response.

[0006] The article, titled "Improved Grid-Connected Control Strategy for VSG Based on Triple Fractional Inertia," published in *Electric Power Automation Equipment*, Vol. 45, No. 07, 2025, pp. 1833-189+224, proposes an improved grid-connected control strategy for VSG based on triple fractional virtual inertia. This strategy can effectively suppress the dynamic oscillation and overshoot of VSG grid-connected active power and output frequency under step disturbances such as active power reference command and grid frequency. It has the advantages of increasing the system's control degrees of freedom and the flexibility of control parameter selection. However, it suffers from problems such as high control structure complexity, parameter tuning dependence on experience, and overshoot in the output frequency response.

[0007] As can be seen from the above, the existing technology can provide some solutions and technical support for suppressing the dynamic oscillation and overshoot of the active power and output frequency of VSG grid connection under disturbances such as active power reference command and grid frequency. However, it still has the disadvantages that the ability to suppress the active power overshoot and output frequency response overshoot of VSG grid connection still needs to be further improved, the increased control freedom is limited, the flexibility of parameter adjustment is limited, and the parameter tuning is difficult. Summary of the Invention

[0008] To overcome the limitations of various technical solutions presented in the background art, this invention addresses the problem of dynamic oscillation and overshoot in the active power and output frequency of traditional virtual synchronous machines under internal and external disturbances such as active power reference commands and grid frequency. It provides a virtual synchronous machine control method based on active power fractional-order high-pass filter feedback. This method optimizes the dynamic response performance of the active power and output frequency of the virtual synchronous machine grid-connected system by introducing a power dynamic compensation stage based on fractional-order high-pass filter feedback into the traditional virtual synchronous machine control structure. This effectively solves the dynamic oscillation and overshoot problems of the active power and output frequency in traditional virtual synchronous machine grid-connected systems under internal and external disturbances such as active power reference commands and grid frequency. Compared to traditional virtual synchronous machines, it has advantages such as more control degrees of freedom, more flexible control parameter selection, and smaller output frequency response overshoot amplitude.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0010] A virtual synchronous machine control method based on active power fractional-order high-pass filter feedback includes the following steps:

[0011] Step 1, power calculation section: First, collect the grid-connected current i of the virtual synchronous machine. a i b i c and output voltage u a ,u b u c The dq component I of the grid-connected current, oriented based on the output phase angle θ of the virtual synchronous machine, is obtained through a single synchronous rotating coordinate transformation. d I q and the dq component of the output voltage U d U q Then, the grid-connected active power P of the virtual synchronous machine is obtained through the power calculation equation. e And grid-connected reactive power Q e ;

[0012] Step 2, primary voltage regulation section, based on the grid-connected reactive power Q of the virtual synchronous machine obtained in Step 1. e The reactive power reference instruction Q of the virtual synchronous machine ref Primary voltage regulation coefficient k qThe voltage reference command E0 is used to obtain the dq axis output voltage reference command E of the virtual synchronizer through a voltage adjustment equation. d E q ;

[0013] Step 3: Based on the power dynamic compensation part fed back by the active power fractional-order high-pass filter, according to the grid-connected active power P of the virtual synchronous machine obtained in Step 1... e and the cutoff angular frequency ω of a fractional-order high-pass filter c Fractional differential order α and feedback compensation coefficient k a The power dynamic compensation amount P of the virtual synchronous machine is obtained through a power dynamic compensation stage based on feedback from a fractional-order high-pass filter. a ;

[0014] Step 4, Rotor Motion Equation Part: The active power reference command P of the virtual synchronizer... ref Subtract the grid-connected active power P of the virtual synchronous machine obtained in step 1 e The power dynamic compensation amount P of the virtual synchronous machine obtained in step 3 a The power difference ΔP is obtained. The power difference ΔP, the virtual inertia coefficient J, the rated angular frequency ω0, and the virtual damping coefficient D of the virtual synchronous machine are used to obtain the angular frequency deviation Δω of the virtual synchronous machine through the rotor motion equation.

[0015] Step 5, Output phase angle generation section: Add the rated angular frequency ω0 to the angular frequency deviation Δω of the virtual synchronizer obtained in step 4 to obtain the output angular frequency ω of the virtual synchronizer. Then, perform an integration operation on the output angular frequency ω to obtain the output phase angle θ of the virtual synchronizer.

[0016] Step 6: First, based on the virtual synchronizer output phase angle θ obtained in Step 5, and the virtual synchronizer dq axis output voltage reference command E obtained in Step 2. d E q The three-phase bridge arm voltage modulation signal E is obtained by inverse transformation of the single synchronous rotating coordinates with the output phase angle θ of the virtual synchronous machine as the orientation reference. a E b E c Then, the three-phase bridge arm voltage modulated signal E a E b E c The SVPWM modulation stage generates the drive signals for the switching transistors of the virtual synchronous machine inverter bridge.

[0017] Preferably, the grid-connected active power P in step 1 e The calculation formula used is:

[0018] P e =1.5(U d I d +Uq I q ),

[0019] Grid-connected reactive power Q e The calculation formula used is:

[0020] Q e =1.5(U q I d -U d I q ).

[0021] Preferably, the output voltage reference command E in step 2 d The calculation formula used is:

[0022] E d =E0+(Q ref -Q e )k q ,

[0023] Output voltage reference command E q The calculation formula used is:

[0024] E q =0.

[0025] Preferably, the power dynamic compensation amount P of the virtual synchronizer in step 3 a The calculation formula used is:

[0026]

[0027] In the formula, s is the Laplace operator.

[0028] Preferably, the formula used to calculate the power difference ΔP in step 4 is:

[0029] ΔP=P ref -P e -P a ,

[0030] The formula for calculating the angular frequency deviation Δω is:

[0031]

[0032] In the formula, s is the Laplace operator.

[0033] Preferably, the formula used to calculate the angular frequency ω of the virtual synchronizer output in step 5 is:

[0034] ω = Δω + ω0,

[0035] The formula for calculating the output phase angle θ is:

[0036]

[0037] In the formula, s is the Laplace operator.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] This invention discloses a virtual synchronous machine control method based on feedback from a fractional-order active power high-pass filter. This method introduces a dynamic power compensation stage based on feedback from a fractional-order active power high-pass filter into the traditional virtual synchronous machine control structure, and flexibly adjusts the cutoff angular frequency ω of the fractional-order high-pass filter. c Fractional differential order α and feedback compensation coefficient k a This invention optimizes the dynamic response performance of active power and output frequency in a virtual synchronous machine grid-connected system. It can effectively solve the problems of dynamic oscillation and overshoot in the active power and output frequency of a traditional virtual synchronous machine grid-connected system under internal and external disturbances such as active power reference command and grid frequency. Compared with the traditional virtual synchronous machine, it has the advantages of more control freedom, more flexible selection of control parameters, and smaller output frequency overshoot amplitude. It can be applied to the field of virtual synchronous machine grid-connected control in power electronics technology. Attached Figure Description

[0040] Figure 1 This is a diagram of the virtual synchronous machine grid-connected control structure according to an embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram illustrating the calculation of the dynamic power compensation amount based on the feedback of the active fractional-order high-pass filter in this invention.

[0042] Figure 3 This is an equivalent control structure diagram of the virtual synchronous machine grid-connected active power closed loop according to an embodiment of the present invention.

[0043] Figure 4 This is a schematic diagram of coordinate transformation and modulation according to an embodiment of the present invention.

[0044] Figure 5 This is a Bode plot comparing the virtual synchronizer before and after adopting this invention.

[0045] Figure 6 This is a comparison of the simulation waveforms of the virtual synchronizer before and after adopting this invention.

[0046] Figure 7 This is a comparison of experimental waveforms before and after the virtual synchronizer was implemented using this invention. Detailed Implementation

[0047] The following detailed embodiments will be further described in conjunction with the above-mentioned figures, as follows:

[0048] Please see Figure 1The present invention proposes a virtual synchronous machine control method based on active power fractional-order high-pass filter feedback, comprising the following steps:

[0049] Step 1, power calculation section: First, collect the grid-connected current i of the virtual synchronous machine. a i b i c and output voltage u a u b u c The dq component I of the grid-connected current, oriented based on the output phase angle θ of the virtual synchronous machine, is obtained through a single synchronous rotating coordinate transformation. d I q and the dq component of the output voltage U d U q Then, the grid-connected active power P of the virtual synchronous machine is obtained through the power calculation equation. e And grid-connected reactive power Q e ;

[0050] Among them, grid-connected active power P e The calculation formula used is:

[0051] P e =1.5(U d I d +U q I q ),

[0052] Grid-connected reactive power Q e The calculation formula used is:

[0053] Q e =1.5(U q I d -U d I q ).

[0054] Step 2, primary voltage regulation section, based on the grid-connected reactive power Q of the virtual synchronous machine obtained in Step 1. e The reactive power reference instruction Q of the virtual synchronous machine ref Primary voltage regulation coefficient k q The voltage reference command E0 of the virtual synchronizer is used to obtain the dq axis output voltage reference command E of the virtual synchronizer through a voltage adjustment equation. d E q ;

[0055] Among them, the output voltage reference command E d The calculation formula used is:

[0056] E d =E0+(Q ref -Q e )k q ,

[0057] Output voltage reference command E q The calculation formula used is:

[0058] E q =0.

[0059] Step 3, the power dynamic compensation part based on the feedback of the active fractional-order high-pass filter, such as... Figure 2 As shown, based on the grid-connected active power P of the virtual synchronous machine obtained in step 1... e and the cutoff angular frequency ω of a fractional-order high-pass filter c Fractional differential order α and feedback compensation coefficient K a The power dynamic compensation amount P of the virtual synchronous machine is obtained through a power dynamic compensation stage based on feedback from a fractional-order high-pass filter. a ;

[0060] Among them, the power dynamic compensation amount P of the virtual synchronous machine a The calculation formula used is:

[0061]

[0062] In the formula, s is the Laplace operator.

[0063] Step 4, Rotor Motion Equation Part: The active power reference command P of the virtual synchronizer... ref Subtract the grid-connected active power P of the virtual synchronous machine obtained in step 1 e The power dynamic compensation amount P of the virtual synchronous machine obtained in step 3 a The power difference ΔP is obtained. The power difference ΔP, the virtual inertia coefficient J, the rated angular frequency ω0, and the virtual damping coefficient D of the virtual synchronous machine are used to obtain the angular frequency deviation Δω of the virtual synchronous machine through the rotor motion equation.

[0064] The formula used to calculate the power difference ΔP is as follows:

[0065] ΔP=P ref -P e -P a ,

[0066] The formula for calculating the angular frequency deviation Δω is:

[0067]

[0068] In the formula, s is the Laplace operator.

[0069] Step 5, Output phase angle generation section: Add the rated angular frequency ω0 to the angular frequency deviation Δω of the virtual synchronizer obtained in step 4 to obtain the output angular frequency ω of the virtual synchronizer. Then, perform an integration operation on the output angular frequency ω to obtain the output phase angle θ of the virtual synchronizer.

[0070] The formula used to calculate the output angular frequency ω of the virtual synchronizer is as follows:

[0071] ω = Δω + ω0,

[0072] The formula for calculating the output phase angle θ is:

[0073]

[0074] In the formula, s is the Laplace operator.

[0075] Combination Figure 1 Based on the above calculations, the grid-connected control structure of the virtual synchronous machine and its grid-connected active power closed-loop equivalent control structure diagram can be obtained as follows: Figure 3 As shown. Figure 3 In this context, δ represents the power factor angle of the virtual synchronizing machine; K represents the synchronization voltage coefficient of the virtual synchronizing machine.

[0076] The formula for calculating the power factor angle δ of the virtual synchronizer is as follows:

[0077]

[0078] The formula for calculating the synchronization voltage coefficient K of a virtual synchronizer is as follows:

[0079]

[0080] In the formula, ω g U is the grid angular frequency. g Let X be the grid voltage amplitude, E be the output voltage amplitude of the virtual synchronous machine, X be the line equivalent inductive reactance, and s be the Laplace operator.

[0081] Step 6, as follows Figure 4 As shown, first, based on the virtual synchronizer output phase angle θ obtained in step 5, and the virtual synchronizer dq axis output voltage reference command E obtained in step 2... d E q The three-phase bridge arm voltage modulation signal E is obtained by inverse transformation of the single synchronous rotating coordinates with the output phase angle θ of the virtual synchronous machine as the orientation reference. a E b E c Then, the three-phase bridge arm voltage modulated signal E a E b E c The SVPWM modulation stage generates the drive signals for the switching transistors of the virtual synchronous machine inverter bridge.

[0082] Example

[0083] To verify the control effect of the proposed virtual synchronous machine control method based on active power fractional-order high-pass filter feedback, the proposed virtual synchronous machine (FO-HPFVSG) optimization control method was compared with existing traditional virtual synchronous machine (VSG) control methods (described in the background art as "Grid-connected Improvement Strategy Based on Triple Fractional-Order Virtual Synchronous Machine", Vol. 45, No. 07, 2025, pp. 1833-189+224) and existing virtual synchronous machine (HPFVSG) control methods based on active power high-pass filter differential feedback (described in the background art as "Transient Power Oscillation Suppression Strategy for Virtual Synchronous Generator Considering Overshoot", Vol. 46, No. 11, 2022, pp. 131-141). The main comparison was made in response to the active power reference command P. ref and grid frequency f g Grid-connected active power P under step disturbance conditions e And the dynamic response performance at the output frequency f. Specifically:

[0084] First, the relevant parameters are set. In this embodiment, the relevant parameters in the FO-HPFVSG control method of the present invention are set as follows:

[0085] The virtual synchronous machine has a rated capacity of 100kVA and an active power reference command P. ref It has a power rating of 20 kW, a rated angular frequency ω0 of 314.16 rad / s, and a virtual inertia coefficient J of 8 kg·m. 2 The virtual damping coefficient D is 50.66 J / rad, and the grid voltage amplitude U g The voltage is 311V, and the primary voltage regulation coefficient is k. q 1.4×10 -4 V / var, the virtual synchronous machine output voltage amplitude E is 311V, the line equivalent inductive reactance X is 0.15Ω, that is, the synchronization voltage coefficient K = 1.5U g E / X is 967210. It is worth noting that if the cutoff angular frequency ω of the fractional-order high-pass filter in the FO-HPFVSG is set... c =0, fractional differential order α=0 and feedback compensation coefficient K aIf α = 0, then FO-HPFVSG is equivalent to VSG, indicating that VSG is only a special case of FO-HPFVSGG; if we set the fractional derivative order α = 1 in FO-HPFVSG, then FO-HPFVSG is equivalent to HPFVSG, indicating that HPFVSG is only a special case of FO-HPFVSG. Meanwhile, combining... Figure 3 Through a series of formula derivations, the active power reference command P of the VSG grid-connected system at this time can be obtained. ref To grid-connected active power P e The closed-loop transfer function is The damping ratio of the corresponding VSG grid-connected active power closed-loop control system That is, the VSG grid-connected active power closed-loop control system is an underdamped system, therefore the VSG grid-connected active power P e With output frequency f in active reference command P ref and grid frequency f g Under step disturbance conditions, dynamic oscillations and overshoot will inevitably occur. Furthermore, according to... Figure 3 Through a series of formula derivations, the active power reference command P of the HPFVSG grid-connected system can be obtained. ref To grid-connected active power P e The closed-loop transfer function is And set the cutoff angular frequency of the high-pass filter in G1(s) to ω. c =166.67 rad / s, feedback compensation coefficient set to K a =19.5, virtual inertia coefficient is set to J = 8 kg·m 2 The virtual damping coefficient is set to D = 50.66 J / rad to ensure that HPFVSG and VSG have the same grid-connected active power steady-state response performance, while also suppressing the grid-connected active power P of HPFVSG. e With output frequency f in active reference command P ref and grid frequency f g Dynamic oscillations existing under step disturbances.

[0086] In this embodiment, the cutoff angular frequency ω of the fractional-order high-pass filter will be optimized. c Fractional differential order α and feedback compensation coefficient K a The value of [value] is selected to further optimize the grid-connected active power P of FO-HPFVSG. e With output frequency f in active reference command P ref and grid frequency f g Dynamic response performance under a step disturbance. Similarly, according to Figure 3 Through a series of formula derivations, the active power reference command P of the FO-HPFVSG grid-connected system can be obtained. refTo grid-connected active power P e The closed-loop transfer function is Therefore, the grid-connected active power dynamic response performance of VSG, HPFVSG, and FO-HPFVSG can be directly analyzed by comparing and contrasting the closed-loop transfer functions G(s), G1(s), and G2(s). To further simplify the theoretical analysis process, the fractional derivative order in FO-HPFVSG is set to α = 1.2, and the cutoff angular frequency ω of the fractional high-pass filter is set to... c =316 rad / s, feedback compensation coefficient set to K a =117. The virtual inertia coefficient is set to J = 8 kg·m 2 The virtual damping coefficient is set to D = 50.66 J / rad. Substituting these parameters into the closed-loop transfer functions G2(s), G1(s), and G(s) respectively, we can obtain the active power reference command step disturbance ΔP of the grid-connected active power closed-loop control system of FO-HPFVSG, HPFVSG, and VSG. ref To the grid-connected active power response ΔP e (ΔP e / ΔP ref For a comparison of the Bode plot, see [link / reference]. Figure 5 (a). Similarly, the grid frequency step disturbance Δω of the grid-connected active power closed-loop control system of FO-HPFVSG, HPFVSG and VSG can be obtained respectively. g To the grid-connected active power response ΔP e (ΔP e / Δω g Active reference command step disturbance ΔP ref Output angular frequency response Δω(Δω / ΔP) ref ) and the step disturbance of the power grid frequency Δω g Output angular frequency response Δ ω (Δω / Δω g For comparison with the Bode plot, please refer to the details below. Figure 5 (b) Figure 5 (c) and Figure 5 (d). Among them, Figure 5 In this diagram, VSG represents an existing conventional virtual synchronous machine control method, meaning the curve indicated corresponds to the Bode plot of the conventional virtual synchronous machine control method before the application of this invention; HPFVSG represents an existing virtual synchronous machine control method based on active power high-pass filter differential feedback, meaning the curve indicated corresponds to the Bode plot of the existing virtual synchronous machine control method based on active power high-pass filter differential feedback before the application of this invention; FO-HPFVSG represents the virtual synchronous machine control method based on active power fractional-order high-pass filter feedback proposed in this invention, meaning the curve indicated corresponds to the Bode plot of the virtual synchronous machine control method based on active power fractional-order high-pass filter feedback proposed in this invention.

[0087] according to Figure 5 It can be seen that, firstly, the traditional VSG grid-connected active power closed-loop control system exhibits significant resonance peaks in the low-to-medium frequency range (10–30 rad / s). This phenomenon indicates that the VSG's grid-connected active power P e With the output angular frequency ω in the active reference command P ref and the angular frequency ω of the power grid g Under step disturbance conditions, there are obvious dynamic oscillations and overshoot problems; II. From Figure 5 The frequency domain analysis results in (a) and 5(c) show that, compared to traditional VSGs, existing HPFVSGs, by introducing a differential feedback structure for an active power high-pass filter, do not exhibit a resonance peak on the amplitude-frequency curve. This phenomenon indicates that the grid-connected active power P of HPFVSGs... e With output frequency f in active reference command P ref No dynamic oscillations were observed under step disturbance conditions. Furthermore, the FO-HPFVSG, through the introduction of an active fractional-order high-pass filter feedback structure, did not exhibit resonance peaks on its amplitude-frequency curves. This phenomenon indicates that the grid-connected active power P of the FO-HPFVSG is effectively controlled. e The output frequency f does not exhibit dynamic oscillation under step disturbance conditions of the active power reference command Pref, and, while improving the system phase margin, it more effectively suppresses active power overshoot and output frequency overshoot in the FO-HPFVSG grid-connected system; III. From Figure 5 The frequency domain analysis results in (b) and 5(d) show that, compared to traditional VSGs, existing HPFVSGs, through the introduction of a high-pass filter differential feedback structure in the active power input, exhibit a smoother trend in the amplitude-frequency curve. However, the amplitude still shows a slight increase in the mid-frequency range. This phenomenon indicates that the grid-connected active power P of HPFVSGs... e With the output frequency f at the grid angular frequency ω g There is no dynamic oscillation under step disturbance conditions, but its grid-connected active power P e At the grid angular frequency ω g Under step disturbance conditions, a significant overshoot phenomenon is observed. Simultaneously, the FO-HPFVSG grid-connected system, through the introduction of an active fractional-order high-pass filter feedback structure, shows no resonance peak in its amplitude-frequency curve. This phenomenon indicates that the grid-connected active power P of the FO-HPFVSG is... e With output frequency f in active reference command P ref No dynamic oscillations are observed under step disturbance conditions. It is worth noting that the FO-HPFVSG also has the advantages of more control parameters and more flexible parameter selection; for example, by selecting a reasonable cutoff angular frequency ω of the fractional-order high-pass filter… c Fractional differential order α and feedback compensation coefficient K a The value of is selected to flexibly optimize the grid-connected active power P of FO-HPFVSG. eThe dynamic response performance with respect to the output frequency f will not be listed in detail due to space limitations.

[0088] Based on the above parameter settings, simulation and experimental comparison tests were conducted, as follows:

[0089] The simulation and experimental test conditions were set as follows: at the initial moment, the virtual synchronous machine stably outputs 20kW of grid-connected active power, and the grid frequency f g Maintaining a constant 50Hz, the active power reference command P is available at 4.0s. ref The power grid frequency f jumps from 20kW to 60kW in 7.0s. g The frequency jumps from 50Hz to 49.95Hz.

[0090] Based on the above working conditions, the following results were obtained: Figure 6 and Figure 7 The simulation and experimental test comparison chart shown below, in which, Figure 6 and Figure 7 In this context, VSG represents an existing traditional virtual synchronous machine control method, HPFVSG represents an existing virtual synchronous machine control method based on active power high-pass filter differential feedback, and FO-HPFVSG represents a virtual synchronous machine control method based on active power fractional-order high-pass filter feedback proposed in this invention. Specifically, the curve pointed to by VSG is the test waveform before using this invention, specifically the dynamic response test waveform of the existing traditional virtual synchronous machine control method; the curve pointed to by HPFVSG is the test waveform before using this invention, specifically the dynamic response test waveform of the existing virtual synchronous machine control method based on active power high-pass filter differential feedback; and the curve pointed to by FO-HPFVSG is the test waveform after using this invention, specifically the test waveform of the virtual synchronous machine control method based on active power fractional-order high-pass filter feedback proposed in this invention.

[0091] according to Figure 6 (a) It can be observed that, firstly, the grid-connected active power P corresponding to the FO-HPFVSG proposed in this invention... e There is no dynamic oscillation with the output frequency f, and the corresponding grid-connected active power P of the VSG is already available. e Both the output frequency f and the output frequency exhibit significant dynamic oscillations and overshoot. Secondly, the overshoot amplitude of the output frequency response of the FO-HPFVSG proposed in this invention is 0.023Hz, which is smaller than the 0.098Hz of the existing VSG and the 0.041Hz of the existing HPFVSG. Therefore, it can be seen that the FO-HPFVSG proposed in this invention, compared to the existing VSG and HPFVSG, has a better performance in terms of active power reference command P. ref During the jump from 20kW to 60kW, the grid-connected active power P e Both the output frequency f and the output frequency have no dynamic oscillations and have a smaller output frequency response overshoot.

[0092] And according to Figure 6 (b) It can be observed that, firstly, the grid-connected active power P corresponding to the FO-HPFVSG proposed in this invention... e There is no dynamic oscillation with the output frequency f, and the corresponding grid-connected active power P of the VSG is already available. e Both the output frequency f and the output frequency exhibit significant dynamic oscillations and overshoot, and the corresponding grid-connected active power P of HPFVSG is already available. e There is significant power overshoot and significant frequency overshoot in the output frequency f; secondly, the grid-connected active power overshoot of the FO-HPFVSG proposed in this invention is 0.78%, which is less than the 17.9% of the existing VSG and the 4.3% of the existing HPFVSG. Therefore, it can be seen that the grid-connected active power overshoot of the FO-HPFVSG proposed in this invention is significantly lower than that of the existing VSG and HPFVSG. e At grid frequency f g Grid-connected active power P during the jump from 50Hz to 49.95Hz e Both the output frequency f and the output frequency f exhibit no dynamic oscillation and have a smaller grid-connected active power overshoot.

[0093] according to Figure 7 (a) It can be found that, firstly, the grid-connected active power P corresponding to the FO-HPFVSG proposed in this invention e There is no dynamic oscillation with the output frequency f, and the corresponding grid-connected active power P of the VSG is already available. e The first issue is that the FO-HPFVSG proposed in this invention exhibits significant dynamic oscillations and power overshoot with respect to the output frequency f. Secondly, the overshoot value of the output frequency response of the FO-HPFVSG proposed in this invention is 0.025Hz, which is smaller than the 0.102Hz of the existing VSG and the 0.044Hz of the existing HPFVSG. Therefore, it can be concluded that the FO-HPFVSG proposed in this invention, compared to the existing VSG and HPFVSG, has a better performance in terms of active power reference command P. ref During the jump from 20kW to 60kW, the grid-connected active power P e Both the output frequency f and the output frequency have no dynamic oscillations and have a smaller output frequency response overshoot.

[0094] And according to Figure 7 (b) It can be found that, firstly, the grid-connected active power P corresponding to the FO-HPFVSG proposed in this invention... e There is no dynamic oscillation with the output frequency f, and the corresponding grid-connected active power P of the VSG is already available. e Both the output frequency f and the output frequency exhibit significant dynamic oscillations and overshoot, and the corresponding grid-connected active power P of HPFVSG is already available. eThere is significant power overshoot and significant frequency overshoot in the output frequency f; secondly, the grid-connected active power overshoot of the FO-HPFVSG proposed in this invention is 0.81%, which is less than the 18.2% of the existing VSG and the 4.5% of the existing HPFVSG. Therefore, it can be seen that the grid-connected active power overshoot of the FO-HPFVSG proposed in this invention is significantly lower than that of the existing VSG and HPFVSG. e At grid frequency f g Grid-connected active power P during the jump from 50Hz to 49.95Hz e Both the output frequency f and the output frequency f exhibit no dynamic oscillation and have a smaller grid-connected active power overshoot.

[0095] Depend on Figure 7 (a) and Figure 6 As can be seen from the comparison of results in (a), in this embodiment... Figure 7 (a) Experimental test results and Figure 6 (a) The simulation test results show a one-to-one correspondence, and both results fully demonstrate that, compared with existing VSGs and existing HPFVSGs, the FO-HPFVSG proposed in this invention has a higher performance in active power reference command P. ref Grid-connected active power P under step disturbance e Since there is no dynamic oscillation at the output frequency f and the output frequency response overshoot is smaller, the FO-HPFVSG proposed in this invention has better control performance compared to existing VSGs and HPFVSGs.

[0096] And again Figure 7 (b) and Figure 6 After comparing the results in (b), it is not difficult to see that in this embodiment... Figure 7 (b) Experimental test results and Figure 6 (b) The simulation test results show a one-to-one correspondence, and both results fully demonstrate that, compared with existing VSGs and existing HPFVSGs, the FO-HPFVSG proposed in this invention has a higher performance at the grid frequency f. g Grid-connected active power P under step disturbance e Since there is no dynamic oscillation with the output frequency f and there is a smaller grid-connected active power overshoot, the FO-HPFVSG proposed in this invention has better control performance compared with existing VSG and HPFVSG.

[0097] The above description is a detailed description of the preferred embodiments of the present invention. However, the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications made under the technical spirit of the present invention should fall within the patent scope covered by the present invention.

Claims

1. A virtual synchronous machine control method based on active fractional-order high-pass filter feedback, characterized in that, Includes the following steps: Step 1, Power Calculation Section: Acquire the grid-connected current i of the virtual synchronous machine. a i b i c and output voltage u a u b u c The dq component I of the virtual synchronous machine grid-connected current is obtained by single synchronous rotating coordinate transformation with the output phase angle θ of the virtual synchronous machine as the orientation reference. d I q and the dq component of the output voltage U d U q Then, the grid-connected active power P of the virtual synchronous machine is obtained through the power calculation equation. e And grid-connected reactive power Q e ; Step 2, primary voltage regulation section, based on the grid-connected reactive power Q of the virtual synchronous machine obtained in Step 1. e The reactive power reference instruction Q of the virtual synchronous machine ref Primary voltage regulation coefficient k q The voltage reference command E0 is used to obtain the dq axis output voltage reference command E of the virtual synchronizer through a voltage adjustment equation. d E q ; Step 3: Based on the power dynamic compensation part fed back by the active power fractional-order high-pass filter, according to the grid-connected active power P of the virtual synchronous machine obtained in Step 1... e and the cutoff angular frequency ω of a fractional-order high-pass filter c Fractional derivative order a, and feedback compensation coefficient k a The power dynamic compensation amount P of the virtual synchronous machine is obtained through a power dynamic compensation stage based on feedback from a fractional-order high-pass filter. a ; Step 4, Rotor Motion Equation Part: The active power reference command P of the virtual synchronizer... ref Subtract the grid-connected active power P of the virtual synchronous machine obtained in step 1 e The power dynamic compensation amount P of the virtual synchronous machine obtained in step 3 a The power difference ΔP is obtained. The power difference ΔP, the virtual inertia coefficient J, the rated angular frequency ω0, and the virtual damping coefficient D of the virtual synchronous machine are used to obtain the angular frequency deviation Δω of the virtual synchronous machine through the rotor motion equation. Step 5, Output phase angle generation section: Add the rated angular frequency ω0 to the angular frequency deviation Δω of the virtual synchronizer obtained in step 4 to obtain the output angular frequency ω of the virtual synchronizer. Then, perform an integration operation on the output angular frequency ω to obtain the output phase angle θ of the virtual synchronizer. Step 6: First, based on the virtual synchronizer output phase angle θ obtained in Step 5, and the virtual synchronizer dq axis output voltage reference command E obtained in Step 2. d E q The three-phase bridge arm voltage modulation signal E is obtained by inverse transformation of the single synchronous rotating coordinates with the output phase angle θ of the virtual synchronous machine as the orientation reference. a E b E c Then, the three-phase bridge arm voltage modulated signal E a E b E c The SVPWM modulation stage generates the drive signals for the switching transistors of the virtual synchronous machine inverter bridge.

2. The virtual synchronous machine control method based on active power fractional-order high-pass filter feedback according to claim 1, characterized in that, The grid-connected active power P in step 1 e The calculation formula used is: P e =1.5(U d AND d +U q AND q ), Grid-connected reactive power Q e The calculation formula used is: Q e =1.5(U q I d -U d I q )。 3. The virtual synchronous machine control method based on active fractional-order high-pass filter feedback according to claim 1, characterized in that, Output voltage reference command E in step 2 d The calculation formula used is: E d <E0+(Q ref -Q e )k q , Output voltage reference command E q The calculation formula used is: E q ＝0。 4. The virtual synchronous machine control method based on active fractional-order high-pass filter feedback according to claim 1, characterized in that, The power dynamic compensation amount P of the virtual synchronizer in step 3 a The calculation formula used is: In the formula, s is the Laplace operator.

5. The virtual synchronous machine control method based on active fractional-order high-pass filter feedback according to claim 1, characterized in that, The formula used to calculate the power difference ΔP in step 4 is: ΔP=P ref -P e -P a , The formula for calculating the angular frequency deviation Δω is: In the formula, s is the Laplace operator.

6. The virtual synchronous machine control method based on active fractional-order high-pass filter feedback according to claim 1, characterized in that, The formula used to calculate the angular frequency ω of the virtual synchronizer output in step 5 is: ω = Δω + ω0, The formula for calculating the output phase angle θ is: In the formula, s is the Laplace operator.

Images