Self-adaptive active support method of grid-connected converter for different resistance-inductance ratios of power grid

Abstract

The invention discloses a grid-connected converter self-adaptive active support method for different resistance-inductance ratios of a power grid, which comprises the following steps of: firstly, constructing a three-phase disturbance source injection grid-connected point, measuring disturbance response of voltage and current of the power grid, and calculating impedance and resistance-inductance ratios of the power grid; secondly, on the basis of power grid impedance, constructing a reference coupling coefficient matrix set containing cross and disturbance groups, and solving a reference coupling coefficient matrix with the highest coefficient fitness; then establishing small signal models of active power and reactive power, and calculating a per-unit coupling coefficient matrix for evaluating a coupling relationship; and finally, performing Hadamard product operation on the reference coupling coefficient matrix and the per-unit coupling coefficient matrix to obtain an actual coupling coefficient matrix, and designing a frequency and voltage control equation of the grid-connected converter with active and reactive components. The grid-connected converter active supporting method is suitable for different power grid resistance-inductance ratios, controller parameters can be adjusted in a self-adaptive mode according to dynamic changes of power grid impedance, and active supporting of frequency and voltage is achieved under various working conditions.

Description

Technical Field

[0001] The present invention belongs to the field of new energy grid-connected power generation, and particularly relates to an adaptive active support method for grid-connected converters facing different resistance-inductance ratios of the power grid. Background Technique

[0002] With the rapid development of renewable energy, especially the wide application of wind energy and photovoltaic power generation, the proportion of distributed power generation in the power grid is gradually increasing. These distributed power sources are usually connected to the power grid through grid-connected converters. The grid-connected converters not only need to convert the direct current of the power source into alternating current synchronized with the power grid, but also must ensure the stability of the power grid and provide voltage and frequency support capabilities. Grid-connected converters are mainly divided into grid-following converters and grid-forming converters. Among them, grid-forming converters have become a current research hotspot because they can actively control voltage and frequency, provide reactive power and short-term dynamic support required by the power grid, and improve the stability of the power grid.

[0003] The control methods of grid-forming grid-connected converters mainly include virtual synchronous machine control and droop control. Virtual synchronous machine control can provide an inertial response similar to that of a physical synchronous machine and improve the stability of the power grid under frequency disturbances. However, the control logic is complex and the dynamic response speed is slow; the principle of droop control is simple, without complex models or algorithms, which is convenient for engineering implementation and applicable to different types of distributed power sources. However, this control method defaults that the inductance of the power grid impedance is much larger than the resistance, making the frequency only related to the active power and the voltage only related to the reactive power, and cannot accurately model the active-reactive coupling relationship.

[0004] The existing active support methods for grid-connected converters have the following problems:

[0005] (1) The current grid-connected converter control method defaults that the inductance of the power grid impedance is much larger than the resistance, assuming that the active power and reactive power are independently decoupled, and ignoring the coupling relationship between frequency, voltage, active power and reactive power caused by the power grid impedance. This control method is not applicable to the current complex grid conditions. When the resistance of the power grid increases, using this method may lead to poor system dynamic response, increased voltage fluctuations, and even cause system oscillation or stability problems;

[0006] (2) The existing control methods are usually based on idealized assumptions, that is, the power grid impedance remains constant and the controller coefficients remain unchanged. However, in the actual power grid, the power grid impedance has dynamic characteristics due to factors such as load fluctuations, line switching, or other power sources being connected to the grid. This makes the output power of the converter deviate greatly from the expected value and even leads to instability, especially in weak power grid or long line environments.

[0007] (3) Many active support methods pay more attention to considering steady-state performance such as the recovery ability of frequency and voltage, and insufficiently consider the balance between dynamic performance and stability margin. Summary of the Invention

[0008] In view of the problems existing in the prior art, the present invention provides an adaptive active support method for grid-connected converters facing different resistance-inductance ratios of the power grid, which can be applied to the current complex grid conditions.

[0009] To solve the above technical problems, the present invention provides the following technical solutions: An adaptive active support method for grid-connected converters facing different resistance-inductance ratios of the power grid, comprising the following steps:

[0010] S1. Construct a three-phase current disturbance source, inject it at the grid connection point when the grid voltage passes through the zero point, then measure the electrical quantities of the grid voltage and current, and input the electrical quantities of the grid voltage and current into a two-stage series complex coefficient filter to obtain the disturbance responses of the voltage and current electrical quantities;

[0011] S2. Divide the filtered voltage by the current to obtain the grid impedance, and further divide the grid reactance by the grid resistance to obtain the resistance-inductance ratio of the grid;

[0012] S3. Based on the reactance and resistance of the grid in S2 respectively, construct a set of reference coupling coefficient matrices containing cross and disturbance groups, obtain the two groups of reference coupling coefficient matrices with the highest coefficient fitness, and merge them into a reference coupling coefficient matrix corresponding to the current grid impedance

[0013] S4. According to the voltage and current of the input grid, obtain the expressions of the active power and reactive power of the input grid, perform small-signal linearization on this expression near the steady-state operating point, and then rewrite it in matrix form to obtain a coupling coefficient matrix H' describing the relationship between frequency, voltage, active power and reactive power PQ ;

[0014] S5. Perform a Hadamard product operation on the coupling coefficient matrix H' PQ and the transpose of its inverse matrix ((H' PQ )) -1 )) T to obtain a normalized coupling coefficient matrix Λ'(H PQ ), and then perform a Hadamard product operation on the reference coupling coefficient matrix in S3 and the normalized coupling coefficient matrix Λ'(H PQ ) to obtain an actual coupling coefficient matrix H;

[0015] S6. Design the grid-connected inverter frequency and voltage control equations with both active and reactive components: The column matrix composed of frequency and voltage is equal to the column matrix composed of rated frequency and rated voltage, plus the column matrix composed of active deviation and reactive deviation multiplied by the actual coupling coefficient matrix H on the left, and finally construct the equation of the grid-connected inverter controller. Combine the output angular frequency and output voltage of the controller to synthesize the voltage phasor, and use the voltage phasor as the reference voltage of the voltage loop controller for subsequent grid-connected inverter control.

[0016] Further, the aforementioned step S1 includes the following sub-steps:

[0017] S101. According to the switching frequency of the grid-connected inverter and the dead-time setting, perform Fourier analysis on the deviation voltage generated by the dead-time to obtain the set of background harmonic frequencies of the inverter: In the formula, f 1 is the fundamental frequency, i is a constant, n is the number of background harmonic frequencies, and f vsci is the i-th background harmonic frequency of the inverter;

[0018] S102. Perform Fourier analysis on the voltage and current components at the grid connection point to obtain the grid background harmonics generated by the nonlinear load, and obtain the set of background harmonic frequencies of the three-phase grid: Among them, is the i-th background harmonic frequency of the three-phase grid;

[0019] S103. Set that the set of harmonic frequencies that can be injected into the grid connection point intersects with the set of inverter background harmonic frequencies to be an empty set, and intersects with the set of grid background harmonic frequencies to be an empty set: In the formula, the set Q represents the current frequency that can be injected into the grid connection point, and Φ represents an empty set;

[0020] S104. Select the frequency f disq in the set Q, take I disq as the amplitude, and θ disq as the phase angle, construct a three-phase current disturbance source, and use the parallel injection method to inject it into the grid connection point when the grid voltage passes through zero: S105. Measure the electrical quantities of the grid voltage and current after injecting the three-phase current disturbance source, convert the two electrical quantities to the αβ coordinate system, and use the α-axis component of the electrical quantity in the αβ coordinate system as the real part and the β-axis component as the imaginary part to construct a complex electrical quantity, and obtain the voltage complex vector U αβ (s), current complex vector I αβ (s), and uniformly represent them with the complex vector X αβ (s)

[0021] X αβ (s) = X α (s) + jX β (s;

[0022] S106. Input the complex vector X αβ (s) into the multi-complex coefficient filter G MCCF (s) of the pre-stage undisturbed decoupling module,

[0023]

[0024] where P i+ and P i- represent the first-order filters corresponding to different background harmonic frequencies: ω i represents the angular frequency of the i-th harmonic, and ω ic is the cut-off frequency of the corresponding first-order complex coefficient filter;

[0025] S107. Subtract the X' αβ (s) output by the pre-stage filter from the initial response X αβ (s) of the voltage or current. The transfer function from the initial response to the filtered response is 1 - G MCCF (s). Calculate the disturbance responses of the voltage and current as follows:

[0026] X' αβq (s) = (1 - G MCCF (s))X αβ (s)

[0027] S108. Input the disturbance responses X' αβq (s) of the voltage and current at the disturbance frequency into the post-stage second-order complex coefficient filters G q+ (s) and G q- (s) to separate the positive-sequence X' αβq (s) and negative-sequence component X' αβq+ (s) of X' αβq- (s) as follows:

[0028]

[0029] where ω q is the angular frequency of the disturbance source ω q = 2πf q , and ω qc is the cut-off frequency of the second-order complex coefficient filter;

[0030] S109. The disturbance source injected into the connection point is a three-phase symmetrical disturbance source, and the negative-sequence component response is zero. Retain the positive-sequence component responses of the filtered voltage and current as follows:

[0031]

[0032] Further, in the aforementioned step S2, the specific steps for calculating the reactance and resistance of the power grid are as follows:

[0033] S201. Divide the filtered voltage by the current to obtain the power grid impedance Z(jω q ), as shown in the following formula:

[0034]

[0035] S202. Divide the reactance of the power grid by the resistance of the power grid to obtain the resistance-inductance ratio X / R of the power grid, as shown in the following formula:

[0036] X / R = X q (jω q ) / R q = X q / R q ;

[0037] In the formula, X q is the reactance of the power grid, and R q is the resistance of the power grid;

[0038] Further, the aforementioned step S3 includes the following sub-steps:

[0039] S301. Based on the reactance X q of the power grid in step S201, randomly generate N groups of initial reference coupling coefficients, each group including an active-frequency coupling coefficient and a reference reactive-voltage coupling coefficient. The initial value range is set as K PW ∈ [a 1 , b 1 , K QV ∈ [a 2 , b 2 . Input each group of reference coupling coefficients into the established Simulink simulation model, measure the dynamic performance and stability margin, and input the comprehensive objective function to calculate the fitness value f(K PWi , K QVi ):

[0040] f(K PWi , K QVi ) = α 1 f dyn (K PWi , K QVi ) + α 2 f stab (K PWi , K QVi )

[0041] Comprehensive objective function: In the formula, f dyn (K PWi , K QVi) is the dynamic performance score of the system when the current coupling coefficient is used. The lower the better the performance, including the adjustment time t s and overshoot σ%, f stab (K PWi , K QVi ) is the stability margin score of the system when the current coupling coefficient is used, GM is the gain margin, which maintains the maximum value of stability, PM is the phase margin, which maintains the minimum phase angle of stability, and GM min 、PM min is the threshold of stability margin, α 1 , α 2 ,ω s ,ω σ％ ,ω GM ,ω PM is the weight coefficient, α 1 +α 2 =1,ω s +ω σ％ =1,ω GM +ω PM =1;

[0042] S302, substitute N groups of benchmark coupling coefficients into the comprehensive objective function to calculate the fitness f(K PWi , K QVi ), retain the first N / 2 groups of benchmark coupling coefficients with high fitness;

[0043] S303, designing an adaptive crossover probability P c : In the formula and are the lower and upper limits of the crossover probability, σ f is the standard deviation of the fitness of the N / 2 groups of benchmark coupling coefficients, σ 0 is a constant, α is a control parameter;

[0044] According to the crossover probability P c Randomly select two sets of reference coupling coefficients (K PWa , K QVa )、(K PWb , K QVb ), and use linear combination to cross-generate two new sets of benchmark coupling coefficients:

[0045]

[0046] In the formula, η∈[0, 1] is the random crossover factor. This step can obtain new N / 2 groups of benchmark coupling coefficients.

[0047] S304. Design an adaptive disturbance probability P m : In the formula and are the lower and upper limits of the perturbation probability, respectively, and β is the control parameter;

[0048] According to the perturbation probability P m randomly select a group of reference coupling coefficients (K PWc , K QVc ) in step S302 to apply perturbations and generate a new group of reference coupling coefficients:

[0049]

[0050] where δ 1 ∈[a 1 , b 1 , δ 2 ∈[a 2 , b 2 is the random perturbation factor, and another new N / 2 groups of reference coupling coefficients are obtained in step S304;

[0051] S305. Combine the reference coupling coefficients in steps S303 and S304 to obtain a new N groups of reference coupling coefficients. Repeat steps S302 - S304 until the maximum number of iterations T is reached, stop the iteration, and output the group of reference coupling coefficients with the highest fitness among the current N groups of reference coupling coefficients

[0052] S306. Calculate the optimal reference reactive - frequency coupling coefficient and reference active - voltage coupling coefficient corresponding to the resistance of the power grid in step S301 according to steps S301 to S305;

[0053] S307. Combine the optimal reference active - frequency coupling coefficient and reference reactive - voltage coupling coefficient in step S305 with the optimal reference reactive - frequency coupling coefficient and reference active - voltage coupling coefficient in step S306 to construct a reference coupling coefficient matrix

[0054] Furthermore, the aforementioned step S4 includes the following sub - steps:

[0055] S401. The current from the grid - connected point to the power grid is: According to the three - phase circuit active power and reactive power calculation formulas P = Re(VI * ), Q = Im(VI * ),

[0056] calculate the active power as follows:

[0057]

[0058] calculate the reactive power as follows:

[0059]

[0060] In the formula, V pcc is the amplitude of the grid connection point voltage, V g is the amplitude of the grid voltage, δ is related to the power angle and frequency, θ q is the impedance angle corresponding to the grid impedance, and the relationship with the short-circuit ratio is θ q = arctan(X q / R q );

[0061] S402. Perform small-signal linearization on the active and reactive power expressions near the steady-state operating point, and perform Taylor expansion on the active power expression and reactive power expression in step S401 near the steady-state operating point (V pcc , θ q ), and retain the first-order terms:

[0062]

[0063] Among them, the partial derivatives are:

[0064]

[0065] Arrange the four partial derivatives into matrix form to obtain the coupling coefficient matrix H PQ :

[0066]

[0067] S403. Combine the conditions of V pcc ≈ V g and the power angle being zero at steady state to simplify the coupling coefficient matrix H' PQ , as follows:

[0068]

[0069] Furthermore, the aforementioned step S5 includes the following sub-steps:

[0070] S501. Take the inverse matrix of the coupling coefficient matrix H' PQ and then perform the transpose operation to obtain ((H' PQ )) - ): 1T :

[0071] S502. Perform the Hadamard product calculation on the coupling coefficient matrix H' PQ and the transpose matrix of its inverse matrix ((H' PQ )) -1 ) T to obtain the per-unit coupling coefficient matrix Λ'(HPQ ):

[0072]

[0073] Wherein, is the Hadamard product, and the operation it represents is the product of the corresponding elements of the matrix, is the per-unit coupling coefficient of active power - frequency, is the per-unit coupling coefficient of reactive power - frequency, is the per-unit coupling coefficient of reactive power - voltage, is the per-unit coupling coefficient of active power - voltage;

[0074] S503. Perform the Hadamard product operation on the per-unit coupling coefficient matrix Λ'(H PQ ) and the reference coupling coefficient matrix in step 307 to obtain the actual coupling coefficient matrix H:

[0075]

[0076] Furthermore, the aforementioned step S6 includes the following sub-steps:

[0077] S601. According to the actual coupling coefficient matrix H in step S503, design the frequency and voltage control equations of the grid-connected converter that takes into account both active and reactive components: The column matrix composed of frequency and voltage is equal to the column matrix composed of rated frequency and rated voltage, plus the column matrix composed of active power deviation and reactive power deviation multiplied by the actual coupling coefficient matrix H on the left:

[0078]

[0079] Wherein, = ω ref is the rated angular frequency, U ref is the rated voltage, P ref is the rated active power, Q ref is the rated reactive power, P is the actual active power of the power grid, and Q is the actual reactive power of the power grid;

[0080] The specific equation of the grid-connected converter controller is:

[0081]

[0082] S602. Combine the output angular frequency and output voltage of the controller to synthesize a voltage phasor, and use it as the reference voltage of the voltage loop controller for subsequent grid-connected converter control.

[0083] Compared with the prior art, the beneficial technical effects of the present invention adopting the above technical solutions are as follows:

[0084] (1) For different resistance-inductance ratios of the power grid, the coupling coefficient matrix can be adaptively adjusted to effectively reduce the influence of the change in the resistance-inductance ratio of the power grid on the control performance, ensuring that the converter can achieve stable operation under weak power grids and different resistance-inductance ratio scenarios;

[0085] (2) This control method can capture the dynamic coupling characteristics of the power grid through an accurate model, further optimize the controller design, enhance the dynamic response ability of the system to disturbances, effectively reduce the control deviation caused by model errors, and improve the steady-state performance;

[0086] (3) This control method not only improves the support capabilities for frequency and voltage, but also optimizes the dynamic performance and stability margin through an optimization algorithm, enabling the system to achieve an optimal balance among different performance indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 is a schematic diagram of the overall process of the present invention;

[0088] Figure 2 is a control block diagram of the present invention;

[0089] Figure 3 is the relationship between the per-unit coupling coefficient and the resistance-inductance ratio in the present invention;

[0090] Figure 4 is a waveform diagram of frequency support under different resistance-inductance ratios in the present invention;

[0091] Figure 5 is a waveform diagram of voltage support under different resistance-inductance ratios in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0092] To better understand the technical content of the present invention, specific embodiments are given below in conjunction with the accompanying drawings for illustration.

[0093] In the present invention, various aspects of the present invention are described with reference to the accompanying drawings, in which many illustrative embodiments are shown. The embodiments of the present invention are not limited to those described in the drawings. It should be understood that the present invention can be implemented by any of the various concepts and embodiments introduced above, as well as the concepts and embodiments described in detail below, because the concepts and embodiments disclosed in the present invention are not limited to any embodiment. Additionally, some aspects disclosed in the present invention can be used alone, or in any suitable combination with other aspects disclosed in the present invention.

[0094] Referring to Figure 1 , the present invention provides an adaptive active support method for grid-connected converters facing different resistance-inductance ratios of the power grid, including the following steps:

[0095] S1. Construct a three-phase current disturbance source, inject it into the grid connection point when the grid voltage passes through zero, then measure the electrical quantities of the grid voltage and current, and input the electrical quantities of the grid voltage and current into a two-stage series complex coefficient filter to obtain the disturbance response of the electrical quantities of the voltage and current;

[0096] S2. Divide the filtered voltage by the current to obtain the grid impedance, and further divide the grid reactance by the grid resistance to obtain the resistance-inductance ratio of the grid;

[0097] S3. Respectively based on the reactance and resistance of the grid in S2, construct a set of reference coupling coefficient matrices including cross and disturbance groups, obtain the two groups of reference coupling coefficient matrices with the highest coefficient fitness, and merge them into the reference coupling coefficient matrix corresponding to the current grid impedance

[0098] S4. According to the voltage and current of the input grid, obtain the expressions of the active power and reactive power of the input grid, perform small-signal linearization on this expression near the steady-state operating point, and then rewrite it in matrix form to obtain the coupling coefficient matrix H' describing the relationship between frequency, voltage, active power and reactive power PQ ;

[0099] S5. Perform the Hadamard product operation on the coupling coefficient matrix H' PQ and the transpose of its inverse matrix ((H ′ PQ ) -1 ) T to obtain the per-unit coupling coefficient matrix Λ'(H PQ ), and then perform the Hadamard product operation on the reference coupling coefficient matrix in S3 and the per-unit coupling coefficient matrix Λ'(H PQ ) to obtain the actual coupling coefficient matrix H;

[0100] S6. Design the grid-connected converter frequency and voltage control equations including both active and reactive components: the column matrix composed of frequency and voltage is equal to the column matrix composed of rated frequency and rated voltage, superimpose the column matrix composed of active power deviation and reactive power deviation and left multiply the actual coupling coefficient matrix H, and finally construct the equation of the grid-connected converter controller, combine the output angular frequency and output voltage of the controller, synthesize the voltage phasor, and use the voltage phasor as the reference voltage of the voltage loop controller for subsequent grid-connected converter control.

[0101] As a preferred embodiment of the present invention, step S1 includes the following sub-steps:

[0102] S101. According to the switching frequency of the grid-connected inverter and the setting of the dead zone, perform Fourier analysis on the deviation voltage generated by the dead zone time to obtain the set of background harmonic frequencies of the inverter:

[0103] In the formula, f 1 is the fundamental frequency, i is a constant, n is the number of background harmonic frequencies, and f vsci is the i-th inverter background harmonic frequency;

[0104] S102. Perform Fourier analysis on the voltage and current components at the grid connection point to obtain the grid background harmonics generated by the nonlinear load, and obtain the set of background harmonic frequencies of the three-phase grid: Among them, is the i-th three-phase grid background harmonic frequency;

[0105] S103. Set that the set of harmonic frequencies that can be injected into the grid connection point intersects with the set of inverter background harmonic frequencies to be an empty set, and intersects with the set of grid background harmonic frequencies to be an empty set: In the formula, the set Q represents the current frequency that can be injected into the grid connection point, and Φ represents an empty set;

[0106] S104. Select the frequency f disq in the set Q, take I disq as the amplitude, and θ disq as the phase angle to construct a three-phase current disturbance source, and use the parallel injection method to inject it into the grid connection point when the grid voltage passes through zero: S105. Measure the electrical quantities of the grid voltage and current after injecting the three-phase current disturbance source, convert the two electrical quantities to the αβ coordinate system, and use the α-axis component of the electrical quantity in the αβ coordinate system as the real part and the β-axis component as the imaginary part to construct a complex electrical quantity, and obtain the voltage complex vector U αβ (s), current complex vector I αβ (s), and uniformly use the complex vector X αβ (s) to represent X αβ (s)=X α (s)+jX β (s;

[0107] S106. Input the complex vector X αβ (s) into the multi-complex coefficient filter G MCCF (s) of the pre-stage non-disturbance decoupling module,

[0108]

[0109] In the formula, P i+ and P i- represent first-order filters corresponding to different background harmonic frequencies: ω i represents the angular frequency of the i-th harmonic, and ω ic is the cut-off frequency of the corresponding first-order complex coefficient filter;

[0110] S107. The initial response X of the voltage or currentαβ (s) Subtract X' output from the pre-stage filter αβ (s). The transfer function corresponding to the initial response to the filtered response is 1 - G MCCF (s). Calculate the disturbance responses of voltage and current as follows:

[0111] X' αβq (s) = (1 - G MCCF (s))X αβ (s)

[0112] S108. Input the disturbance responses X' αβq (s) of voltage and current at the disturbance frequency into the post-stage second-order complex coefficient filters G q+ (s) and G q- (s), and separate the positive sequence X' αβq (s) and negative sequence component X' αβq+ (s) of X' αβq- (s) as follows:

[0113]

[0114] In the formula, ω q is the angular frequency ω of the disturbance source q = 2πf q , ω qc is the cut-off frequency of the second-order complex coefficient filter;

[0115] S109. The disturbance source at the point of injection is a three-phase symmetrical disturbance source, and the negative sequence component response is zero. Retain the positive sequence component responses of the filtered voltage and current as follows:

[0116]

[0117] As a preferred embodiment of the present invention, in step S2, the specific steps for calculating the reactance and resistance of the power grid are as follows:

[0118] S201. Divide the filtered voltage by the current to obtain the power grid impedance Z(jω q ), as follows:

[0119]

[0120] S202. Divide the power grid reactance by the power grid resistance to obtain the resistance-inductance ratio X / R of the power grid, as follows:

[0121] X / R = X q (jω q ) / R q = X q / R q ;

[0122] In the formula, X q is the reactance of the power grid, and R q is the resistance of the power grid;

[0123] As a preferred embodiment of the present invention, step S3 includes the following sub-steps:

[0124] S301. Based on the reactance X of the power grid in step S201 q , use the MATLAB program to randomly generate N groups of initial reference coupling coefficients. Each group includes an active-power - frequency coupling coefficient and a reference reactive-power - voltage coupling coefficient. The initial value range is set as K PW ∈[a 1 , b 1 , K QV ∈[a 2 , b 2 . Input each group of reference coupling coefficients into the established Simulink simulation model, measure the dynamic performance and stability margin, and input the comprehensive objective function to calculate the fitness value f(K PWi , K QVi ):

[0125] f(K PWi , K QVi ) = α 1 f dyn (K PWi , K QVi ) + α 2 f stab (K PWi , K QVi )

[0126] Comprehensive objective function: In the formula, f dyn (K PWi , K QVi ) is the dynamic performance score of the system when the current coupling coefficient is adopted. The lower the score, the better the performance, including the adjustment time t s and the overshoot σ%. f stab (K PWi , K QVi ) is the stability margin score of the system when the current coupling coefficient is adopted. GM is the gain margin, the maximum value to maintain stability, PM is the phase margin, the minimum phase angle to maintain stability, GM min , PM min are the thresholds of the stability margin, α 1 , α 2 , ω s , ω σ％ , ω GM , ω PM are the weight coefficients, α 1 +α 2 = 1, ωs +ω σ％ = 1, ω GM +ω PM = 1;

[0127] S302. Substitute N groups of benchmark coupling coefficients into the comprehensive objective function to calculate the fitness f(K PWi , K QVi ), and retain the first N / 2 groups of benchmark coupling coefficients with high fitness;

[0128] S303. Design an adaptive crossover probability P c : In the formula and are the lower and upper limits of the crossover probability respectively, σ f is the standard deviation of the fitness of N / 2 groups of benchmark coupling coefficients, σ 0 is a constant, and α is a control parameter;

[0129] Randomly select two groups of benchmark coupling coefficients (K c , K PWa ), (K QVa ), (K PWb , K QVb ) in step 3202 according to the crossover probability P

[0130]

[0131] In the formula, η ∈ [0, 1] is a random crossover factor, and this step can obtain a new N / 2 groups of benchmark coupling coefficients;

[0132] S304. Design an adaptive perturbation probability P m : In the formula and are the lower and upper limits of the perturbation probability respectively, and β is a control parameter;

[0133] Randomly select a group of benchmark coupling coefficients (K m , K PWc ), (K QVc ) in step S302 according to the perturbation probability P

[0134]

[0135] In the formula, δ 1 ∈ [a 1 , b 1 , δ 2 ∈ [a 2 , b 2is the random perturbation factor, and another new N / 2 groups of reference coupling coefficients are obtained in step S304;

[0136] S305. Combine the reference coupling coefficients in step S303 and step S304 to obtain new N groups of reference coupling coefficients. Repeat steps S302 - S304 until the maximum number of iterations T is reached, stop the iteration, and output the group of reference coupling coefficients with the highest fitness among the current N groups of reference coupling coefficients

[0137] S306. Calculate the optimal reference reactive - frequency coupling coefficient and reference active - voltage coupling coefficient corresponding to the resistance of the power grid in step S301 according to steps S301 to S305;

[0138] S307. Combine the optimal reference active - frequency coupling coefficient and reference reactive - voltage coupling coefficient in step S305 with the optimal reference reactive - frequency coupling coefficient and reference active - voltage coupling coefficient in step 306 to construct a reference coupling coefficient matrix

[0139] As a preferred embodiment of the present invention, step S4 includes the following sub - steps:

[0140] S401. The current from the grid connection point to the power grid is: According to the three - phase circuit active power and reactive power calculation formulas P = Re(VI * ), Q = Im(VI * )

[0141] Calculate the active power as follows:

[0142]

[0143] Calculate the reactive power as follows:

[0144]

[0145] In the formula, V pcc is the grid - connection point voltage amplitude, V g is the power grid voltage amplitude, δ is related to the power angle and frequency, θ q is the impedance angle corresponding to the power grid impedance, and its relationship with the short - circuit ratio is θ q = arctan(X q / R q );

[0146] S402. Perform small - signal linearization on the active and reactive power expressions near the steady - state operating point. For the active power expression and reactive power expression in step S401 at the steady - state operating point (V pcc , θ q) Perform a Taylor expansion in the vicinity and retain the first-order term:

[0147]

[0148] Among them, the partial derivatives are:

[0149]

[0150] Arrange the four partial derivatives into matrix form to obtain the coupling coefficient matrix H that describes the relationship between frequency, voltage, active power, and reactive power PQ :

[0151]

[0152] S403. Combine the condition that at steady state, V pcc ≈V g , and the power angle is zero to simplify the coupling coefficient matrix H' PQ , as follows:

[0153]

[0154] As a preferred embodiment of the present invention, step S5 includes the following sub-steps:

[0155] \S501. Take the inverse matrix of the coupling coefficient matrix H' PQ and then perform a transpose operation to obtain ((H' PQ ) -1 ) T :

[0156]

[0157] S502. Perform a Hadamard product calculation on the coupling coefficient matrix H' PQ and the transpose matrix of its inverse matrix ((H' PQ ) -1 ) T to obtain the per-unit coupling coefficient matrix Λ'(H PQ ):

[0158]

[0159] In the formula, is the Hadamard product, and the operation it represents is the product of the corresponding elements of the matrix, is the per-unit active-frequency coupling coefficient, and the larger this value is, the greater the coupling degree between frequency and active power, is the per-unit reactive-frequency coupling coefficient, and the larger this value is, the greater the coupling degree between frequency and reactive power, is the per-unit reactive-voltage coupling coefficient, and the larger this value is, the greater the coupling degree between voltage and reactive power, is the per-unit coupling coefficient of active power and voltage. The larger this value is, the greater the coupling degree between voltage and active power;

[0160] S503. Perform the Hadamard product operation on the per-unit coupling coefficient matrix Λ'(H PQ ) in step S502 and the reference coupling coefficient matrix in step S307 to obtain the actual coupling coefficient matrix H:

[0161]

[0162] As a preferred embodiment of the present invention, step S6 includes the following sub-steps:

[0163] S601. According to the actual coupling coefficient matrix H in step S503, design the frequency and voltage control equations of the grid-connected converter that takes into account both active and reactive components: The column matrix composed of frequency and voltage is equal to the column matrix composed of rated frequency and rated voltage, and the column matrix composed of the active power deviation and reactive power deviation is left-multiplied by the actual coupling coefficient matrix H:

[0164]

[0165] where, = ω ref is the rated angular frequency, U ref is the rated voltage, P ref is the rated active power, Q ref is the rated reactive power, P is the actual active power of the power grid, and Q is the actual reactive power of the power grid;

[0166] The specific equation of the grid-connected converter controller is:

[0167]

[0168] S602. Combine the output angular frequency and output voltage of the controller to synthesize a voltage phasor, and use it as the reference voltage of the voltage loop controller for subsequent grid-connected converter control.

[0169] The following further illustrates this method with the results of specific embodiments. Table 1 gives the key simulation parameters of the grid-connected converter model for different resistance-inductance ratios of the power grid:

[0170] Table 1

[0171]

[0172]

[0173] By injecting a three-phase disturbance source into three power grids, three groups of power grid impedances are measured, Z 1 =(0.0735 + j0.679) Ω, Z 2= (0.106 + j0.532) Ω, Z 3 = (0.142 + j0.498) Ω. Table 2 gives the reference coupling coefficient matrices corresponding to the three impedances, and Table 3 gives the resistance-inductance ratios and per-unit coupling coefficient matrices under three different grid impedances.

[0174] Table 2

[0175]

[0176] Table 3

[0177]

[0178] Figure 2 is the overall control block diagram, and the shaded part is the controller part designed by the present invention; Figure 3 is the per-unit coupling coefficient corresponding to different resistance-inductance ratios, and each grid impedance angle corresponds to a determined per-unit coupling coefficient; Figure 4 is the adaptive support effect of the frequency under three impedance ratios, and it can be seen that the frequency response is fast and there is no overshoot; Figure 5 is the adaptive support effect of the voltage when there are disturbances under three impedance ratios, and it can be seen that the overvoltage amplitude is small, the overall response is rapid, and the final overshoot is small.

[0179] In the description of this specification, the descriptions referring to terms such as "an embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0180] Although the present invention has been described above with preferred embodiments, it is not intended to limit the present invention. Those with ordinary knowledge in the technical field to which the present invention pertains can make various modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to what is defined by the claims.

Claims

1. An adaptive active support method for grid-connected converters with different resistance-inductance ratios for power grids, characterized in that: The following steps are involved: S1. Construct a three-phase current disturbance source, inject it into the grid connection point when the grid voltage passes through zero, then measure the grid voltage and current electrical quantities, and input the grid voltage and current electrical quantities into a two-stage series complex coefficient filter to obtain the disturbance response of the voltage and current electrical quantities; S2, dividing the filtered voltage and current to obtain the grid impedance, and further dividing the grid reactance and grid resistance to obtain the grid resistance-inductance ratio; S3, based on the reactance and resistance of the power grid in S2, respectively, construct a set of reference coupling coefficient matrices containing crossover and disturbance groups, obtain the two sets of reference coupling coefficient matrices with the highest coefficient fitness, and merge them into the reference coupling coefficient matrix corresponding to the current power grid impedance S4. According to the voltage and current of the input grid, the expressions of active power and reactive power of the input grid are obtained. The expressions are linearized near the steady-state operating point and rewritten into matrix form to obtain the coupling coefficient matrix H′ describing the relationship between frequency, voltage, active power and reactive power. PQ ; S5. The coupling coefficient matrix H′ PQ The transpose of its inverse matrix ((H′ PQ ) -1 ) T Perform Hadamard product operation to obtain the per-unit coupling coefficient matrix Λ′ (H PQ ), and then the reference coupling coefficient matrix in S3 and the per-unit coupling coefficient matrix Λ′(H PQ ) performs Hadamard product operation to obtain the actual coupling coefficient matrix H; S6. Design the frequency and voltage control equations of the grid-connected converter with both active and reactive components: the column matrix composed of frequency and voltage is equal to the column matrix composed of rated frequency and rated voltage, and the column matrix composed of superimposed active deviation and reactive deviation is multiplied on the left by the actual coupling coefficient matrix H. Finally, the equation of the grid-connected converter controller is constructed, the output angular frequency of the controller is combined with the output voltage, the voltage phasor is synthesized, and the voltage phasor is used as the reference voltage of the voltage loop controller for subsequent grid-connected converter control.

2. The method for adaptive active support of grid-connected converters with different resistance-inductance ratios for power grids according to claim 1 is characterized in that: Step S1 includes the following sub-steps: S101, according to the switching frequency and dead zone setting of the grid-connected inverter, the background harmonic frequency set of the inverter is obtained by Fourier analysis of the deviation voltage generated by the dead zone time: V: Where f1 is the fundamental frequency, i is a constant, n is the number of background harmonic frequencies, and f vsci is the background harmonic frequency of the ith inverter; S102, perform Fourier analysis on the voltage and current components of the grid connection point to obtain the background harmonics of the grid generated by the nonlinear load, and obtain the background harmonic frequency set of the three-phase grid: G: in, is the i-th three-phase power grid background harmonic frequency; S103, setting the harmonic frequency set that can be injected into the grid connection point to be an empty set when intersecting with the inverter background harmonic frequency set, and setting the harmonic frequency set that intersects with the power grid background harmonic frequency set to be an empty set: In the formula, the set Q represents the current frequency that can be injected into the grid connection point, and Φ represents the empty set; S104. Select frequency f in set Q disq , take I disq is the amplitude, θ disq As the phase angle, construct a three-phase current disturbance source, use the parallel injection method, and inject it into the grid connection point when the grid voltage passes through zero point: S105, measure the voltage and current electrical quantities of the power grid after the three-phase current disturbance source is injected, convert the two electrical quantities into the αβ coordinate system, and use the α-axis component of the electrical quantity in the αβ coordinate system as the real part and the β-axis component as the imaginary part to construct the complex electrical quantity, and obtain the voltage complex vector U αβ (s), current complex vector I αβ (s), unified by complex vector X αβ (s) indicates X αβ (s) = X α (s)+jX β (s); S106, complex vector X αβ (s) Input pre-stage disturbance-free decoupling module multi-complex coefficient filter G MCCF (s), Where P i+ and P i- Representation of first-order filters for different background harmonic frequencies: ω i represents the angular frequency of the i-th harmonic, ω ic is the cutoff frequency of the corresponding first-order complex coefficient filter; S107, the initial response of voltage or current X αβ (s) minus the output of the pre-stage filter X' αβ (s), the transfer function from the initial response to the filtered response is 1-G MCCF (s), calculate the disturbance response of voltage and current as follows: X' αβq (s)=(1-G MCCF (s))X αβ (s) S108, the disturbance response X' of the voltage and current at the disturbance frequency αβq (s) Input post-stage second-order complex coefficient filter G q+ (s) and G q- (s), separate X' αβq (s) positive sequence X' αβq+ (s) and negative sequence component X' αβq- (s), as follows: In the formula, ω q is the angular frequency ω of the disturbance source q =2πf q ,ω qc is the cutoff frequency of the second-order complex coefficient filter; S109, the disturbance source injected into the grid connection point is a three-phase symmetrical disturbance source, the negative sequence component response is zero, and the filtered voltage and current positive sequence component response is retained, as shown in the following formula:

3. The method for adaptive active support of grid-connected converters with different resistance-inductance ratios for power grids according to claim 1 is characterized in that: In step S2, the specific steps for calculating the reactance and resistance of the power grid are as follows: S201, divide the filtered voltage and current to obtain the grid impedance Z(jω q ), as follows: S202. Divide the grid reactance by the grid resistance to obtain the grid resistance-inductance ratio X / R, as shown in the following formula: X / R=X q (jω q ) / R q =X q / R q ; Where, X q is the grid reactance, R q is the grid resistance.

4. The method for adaptive active support of grid-connected converters with different resistance-inductance ratios for power grids according to claim 1 is characterized in that: Step S3 includes the following sub-steps: S301, based on the reactance X of the power grid in step S201 q , randomly generate N groups of initial benchmark coupling coefficients, each group contains active-frequency coupling coefficients and benchmark reactive-voltage coupling coefficients, and the initial value range is set to K PW ∈[a1,b1],K QV ∈[a2, b2], input each set of benchmark coupling coefficients into the built Simulink simulation model, measure the dynamic performance and stability margin, and input the comprehensive objective function to calculate the fitness value f(K PWi , K QVi ): Comprehensive objective function: In the formula, f dyn (K PWi , K QVi ) is the dynamic performance score of the system when the current coupling coefficient is used. The lower the better the performance, including the adjustment time t s and overshoot σ%, f stab (K PWi , K QVi ) is the stability margin score of the system when the current coupling coefficient is used, GM is the gain margin, which maintains the maximum value of stability, PM is the phase margin, which maintains the minimum phase angle of stability, and GM min 、PM min is the threshold value of stability margin, α1, α2, ω s ,ω σ％ ,ω GM ,ω PM is the weight coefficient, α1+α2=1, ω s +ω σ％ =1,ω GM +ω PM =1; S302, substitute N groups of benchmark coupling coefficients into the comprehensive objective function to calculate the fitness f(K PWi , K QVi ), retain the first N / 2 groups of benchmark coupling coefficients with high fitness; S303, design adaptive crossover probability P c : In the formula and are the lower and upper bounds of the crossover probability, σ f is the standard deviation of the fitness of the N / 2 groups of benchmark coupling coefficients, σ0 is a constant, and α is a control parameter; According to the crossover probability P c Randomly select two sets of reference coupling coefficients (K PWa , K QVa )、(K PWb , K QVb ), and use linear combination to cross-generate two new sets of benchmark coupling coefficients: In the formula, η∈[0, 1] is the random crossover factor. This step can obtain new N / 2 groups of benchmark coupling coefficients. S304. Design an adaptive disturbance probability P m : In the formula and are the lower and upper limits of the disturbance probability, and β is the control parameter; According to the perturbation probability P m Randomly select a set of reference coupling coefficients (K) in step S302 PWc , K QVc ) to generate a new set of reference coupling coefficients: Wherein, δ1∈[a1, b1], δ2∈[a2, b2] are random disturbance factors, and step S304 obtains another new N / 2 groups of reference coupling coefficients; S305: Combine the reference coupling coefficients in step S303 and step S304 to obtain new N groups of reference coupling coefficients, repeat steps S302 to S304 until the maximum number of iterations T is reached, stop iteration, and output the group of reference coupling coefficients with the highest fitness among the current N groups of reference coupling coefficients. S306, calculating the optimal reference reactive power-frequency coupling coefficient and reference active power-voltage coupling coefficient corresponding to the resistance of the power grid in step S301 according to steps S301 to S305; S307: Combine the optimal reference active-frequency coupling coefficient and reference reactive-voltage coupling coefficient in step S305 with the optimal reference reactive-frequency coupling coefficient and reference active-voltage coupling coefficient in step S306 to construct a reference coupling coefficient matrix.

5. The method for adaptive active support of grid-connected converters with different resistance-inductance ratios for power grids according to claim 1, characterized in that: Step S4 includes the following sub-steps: S401. The current from the grid connection point to the grid is: According to the three-phase circuit active power and reactive power calculation formula P = Re (VI * ),Q=Im(VI * ), Calculate the active power as follows: Calculate the reactive power as follows: Where V pcc is the voltage amplitude at the grid connection point, V g is the grid voltage amplitude, δ is the power angle related to frequency, θ q is the impedance angle corresponding to the grid impedance, and its relationship with the short-circuit ratio is θ q =arctan(X q / R q ); S402, perform small signal linearization on the active power and reactive power expressions near the steady-state operating point, and perform small signal linearization on the active power expression and reactive power expression of step S401 near the steady-state operating point (V pcc ,θ q ) and retain the first-order terms: The partial derivative is: The four partial derivatives are organized into a matrix form to obtain the coupling coefficient matrix H describing the relationship between frequency, voltage, active power and reactive power PQ : S403, combined with steady state V pcc ≈V g , the power angle is zero conditional simplified coupling coefficient matrix H' PQ , as follows:

6. The method for adaptive active support of grid-connected converters with different resistance-inductance ratios for power grids according to claim 1, characterized in that: Step S5 includes the following sub-steps: S501, coupling coefficient matrix H' PQ Take the inverse matrix and then transpose it to get ((H' PQ ) -1 ) T : S502, the coupling coefficient matrix H' PQ The transposed matrix of its inverse matrix ((H' PQ ) -1 ) T The Hadamard product is calculated to obtain the per-unit coupling coefficient matrix Λ' (H PQ ): In the formula, is the Hadamard product, which means the operation is the product of corresponding elements of the matrix. is the active-frequency per unit coupling coefficient, is the reactive-frequency per unit coupling coefficient, is the reactive-voltage per unit coupling coefficient, is the active-voltage per unit coupling coefficient; S503, the unitary coupling coefficient matrix Λ' (H PQ ) and the reference coupling coefficient matrix in step S307 Perform the Hadamard product operation to obtain the actual coupling coefficient matrix H:

7. The method for adaptive active support of grid-connected converters with different resistance-inductance ratios for power grids according to claim 1, characterized in that: Step S6 includes the following sub-steps: S601. According to the actual coupling coefficient matrix H in step S503, the frequency and voltage control equations of the grid-connected converter with both active and reactive components are designed: the column matrix composed of frequency and voltage is equal to the column matrix composed of rated frequency and rated voltage, and the column matrix composed of superimposed active deviation and reactive deviation is multiplied by the actual coupling coefficient matrix H on the left: In the formula, =ω ref is the rated angular frequency, U ref is the rated voltage, P ref is the rated active power, Q ref is the rated reactive power, P is the actual active power of the power grid, and Q is the actual reactive power of the power grid; The specific equation of the grid-connected converter controller is: S602: Combine the output angular frequency of the controller with the output voltage to synthesize a voltage phasor, and use it as a reference voltage of the voltage loop controller for subsequent grid-connected converter control.

Images