A control strategy for asymmetric fault ride-through of grid-connected virtual synchronous generators

Abstract

This invention discloses a control strategy for asymmetric fault ride-through of a grid-type virtual synchronous generator, belonging to the field of renewable energy power generation. The strategy comprises the following steps: positive and negative sequence component extraction; positive and negative sequence virtual synchronous control; positive and negative sequence virtual impedance and positive and negative sequence voltage loops; current limiting with a current limiter and positive and negative sequence current loops. Compared to existing technologies, this invention enables the virtual synchronous generator to conform to the operating characteristics of a traditional synchronous generator with symmetrical output voltage under asymmetric faults. Furthermore, the electrical angle of the negative sequence loop is generated by the negative sequence power loop, eliminating the steady-state error of the negative sequence control loop.

Description

Technical Field

[0001] The present invention belongs to the field of renewable energy power generation, and in particular relates to an asymmetric fault ride-through control strategy for a grid-connected virtual synchronous generator based on positive and negative sequence separation. Background Art

[0002] As the penetration rate of renewable energy increases, the damping and inertia of the power system itself will be weakened due to the lack of damping and inertia in traditional simple grid-following static converters based on phase-locked loops (PLLs). When the power system experiences an active power imbalance, the rate of change of the grid frequency is relatively high, weakening the grid frequency stability. Severe active power imbalances can even cause the power system frequency to collapse, triggering large-scale power outages. In addition, due to the limitations of the PLL, traditional PLL-based grid-following converters are prone to instability in weak grid conditions. Due to the current source characteristics of the grid-following converter, it cannot spontaneously provide reactive power support during low-voltage grid faults. Instead, it needs to generate reactive power by crossing the reactive power curve through a specific low-voltage fault, which relies on the dynamic response time of the PLL.

[0003] To compensate for the inertia and damping missing from traditional grid-connected converters and to better support the power grid, virtual synchronous generator (VSG) and grid-connecting technologies have emerged. Virtual synchronous generators simulate the operating characteristics of synchronous generators to simulate the required inertia and droop characteristics, participating in the inertial frequency regulation of the power system and facilitating frequency stability. Grid-connecting technology controls a voltage-source grid-connected converter with a given voltage amplitude and frequency. Due to these voltage source characteristics, when a low-voltage fault occurs in the grid, the grid-connecting converter can spontaneously and rapidly provide reactive power support within milliseconds, contributing to voltage stability in the power system. Grid-connecting VSGs have therefore attracted increasing attention. However, despite significant efforts from academia and industry, fault ride-through remains an open research issue for grid-connecting converters, particularly for asymmetric faults, a common type of fault in power grids.

[0004] At present, there are some papers that have studied the control of asymmetric fault ride-through of grid-type virtual synchronous generators. The paper "Wan Xiaofeng, Hu Hailin, Nie Xiaoyi, et al. Improved virtual synchronous generator control strategy under grid voltage imbalance [J]. Power System Technology, 2017, 41(11): 3573-3581" proposed a control scheme that uses quasi-static virtual impedance to replace the voltage loop to balance the current of the virtual synchronous generator under asymmetric fault (or suppress the fluctuation of active and reactive power twice the grid frequency). However, due to the lack of voltage loop, steady-state error may occur in voltage control. Moreover, the control of balancing current (or suppressing the fluctuation of active and reactive power twice the grid frequency) under asymmetric fault does not conform to the output characteristics of traditional synchronous generators under asymmetric fault. Under asymmetric fault, the virtual synchronous generator should output a symmetrical grid-connected voltage to the grid like a synchronous generator. Moreover, this scheme does not consider the fault current limiting of the converter, which may lead to converter failure under severe low voltage fault.

[0005] Document "B.Mahamedi, M.Eskandari, JEFletcher and J.Zhu, "Sequence-BasedControl Strategy With Current Limiting for the Fault Ride-Through ofInverter--Interfaced Distributed Generators," in IEEE Transactions onSustainable Energy, vol.11, no.1, pp.165-174, Jan.2020, doi:10.1109 / TSTE.2018.2887149." ("B. Mahamedi, M. Eskandari, JEFletcher and J. Zhu (2018) proposed a V / f control-based asymmetric fault positive-negative sequence separation and current limiting control strategy for inverter-connected distributed generators. The strategy uses a notch filter to filter out the double-frequency component of the asymmetric three-phase signal after DQ conversion, and then controls the inverter through the positive and negative sequence voltage and current loops. However, the current limiting method uses wave-by-wave phase-by-phase current limiting, which may cause the loss of the grid voltage source characteristic under some severe faults. In addition, the electrical angle of the negative sequence loop is obtained by directly inverting the positive sequence loop, which will lead to steady-state errors in the control of the negative sequence loop.

[0006] In summary, the existing solutions still have the following shortcomings:

[0007] 1) The control target does not conform to the operating characteristics of symmetrical output voltage under asymmetric fault of traditional synchronous generators;

[0008] 2) The method of obtaining the electrical angle of the negative sequence loop is not considered. Instead, the electrical angle of the positive sequence loop is roughly reversed, which will lead to a steady-state error in the negative sequence control loop. Summary of the Invention

[0009] The problem to be solved by the present invention is to overcome the limitations of the above-mentioned scheme and propose a grid-type virtual synchronous generator asymmetric fault ride-through control strategy based on positive and negative sequence separation, which can meet the operating characteristics of the symmetrical output voltage of the traditional synchronous generator under asymmetric faults, and the electrical angle of the negative sequence loop is generated by the negative sequence power loop, which can eliminate the steady-state error of the negative sequence control loop.

[0010] To solve the technical problem of the present invention, the present invention provides a grid-type virtual synchronous generator asymmetric fault ride-through control strategy. The circuit topology involved in the control strategy includes a DC side power supply, an inverter, a three-phase LC filter, a grid impedance and a three-phase grid. The three-phase LC filter includes a three-phase filter inductor and a three-phase filter capacitor; the DC side power supply, inverter, filter inductor, grid impedance and three-phase grid are connected in series in sequence, and the other end of the three-phase grid is grounded; one end of the filter capacitor is connected between the filter inductor and the grid impedance according to the phase sequence, and the other end is grounded;

[0011] The control strategy includes the following steps:

[0012] Step 1: Given the positive sequence active power command value P of the virtual synchronous generator ref1 , positive sequence reactive power command value Q ref1 , and let the negative sequence active power command value P ref2 =0;

[0013] Sampling the inverter side current I abc , grid-side current I gabc and capacitor voltage V abc The following 12 positive and negative sequence dq components are extracted by SOGI second-order generalized integrator, including: inverter side current positive sequence d-axis component I d1 , inverter side current positive sequence q-axis component I q1 , the negative sequence d-axis component of the inverter side current I d2 , the negative sequence q-axis component of the inverter side current I q2 , grid-side current positive sequence d-axis component I gd1 , grid-side current positive sequence q-axis component I gq1 , grid-side current negative sequence d-axis component I gd2 , grid-side current negative sequence q-axis component I gq2 、Capacitor voltage positive sequence d-axis component Vd1 , capacitor voltage positive sequence q-axis component V q1 , capacitor voltage negative sequence d-axis component V d2 and the capacitor voltage negative sequence q-axis component V q2 ;

[0014] Step 2: Calculate the positive sequence active power P according to the 12 dq axis components extracted in step 1. e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 ;

[0015] According to the positive sequence active power P e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 , calculate the positive sequence reference electrical angle θ1, the negative sequence reference electrical angle θ2 and the positive sequence reactive loop output E1 through the virtual synchronous generator control algorithm, and set the negative sequence reactive loop output E2 = 0;

[0016] Step 3: Given a virtual resistor R v and virtual reactance X v , calculate the positive sequence d-axis voltage drop ΔV d1 , positive sequence q-axis voltage drop ΔV q1 , negative sequence d-axis voltage drop ΔV d2 and negative sequence q-axis voltage drop ΔV q2 Then, the positive sequence d-axis voltage loop command value V is calculated based on the four voltage drops. dref1 , positive sequence q axis voltage loop command value V qref1 , negative sequence d-axis voltage loop command value V dref2 and the negative sequence q-axis voltage loop command value V qref2 Then, the first positive sequence d-axis current loop command value I generated by the positive and negative sequence voltage loop is calculated based on the four voltage loop command values. dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2 ;

[0017] Step 4: Correct the coefficient K according to the current command value lim , calculate the second positive sequence d-axis current loop command value I after passing through the positive and negative sequence current loop and current limiting dref1 ', the second positive sequence q-axis current loop command value I qref1 ', the second negative sequence d-axis current loop command value I dref2 ′ and the second negative sequence q-axis current loop command value I qref2 ′, and its calculation formulas are:

[0018]

[0019] Perform positive and negative sequence current loop control and output modulation wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , whose expression is:

[0020]

[0021] Among them, K pi is the current regulator proportional coefficient, K ii is the current regulator integral coefficient;

[0022] The modulated wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , after inverse Clarke transform and inverse Park transform, the positive sequence three-phase modulation wave U is obtained a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 ;

[0023] According to the positive sequence three-phase modulation wave U a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 Get the three-phase modulation wave U a , U b , U c , whose expression is:

[0024]

[0025] Preferably, the steps of extracting 12 positive and negative sequence dq components by using a SOGI second-order generalized integrator in step 1 are as follows:

[0026] Introducing three-phase signal S abc ;

[0027] First, the three-phase signal S is transformed by Clarke abc Transformed into αβ coordinate system signal S a , S β , its transformation formula is:

[0028]

[0029] Then the αβ coordinate system signal S a , S β It is sent to the SOGI second-order generalized integrator to generate the orthogonal signal S α e -jπ / 2 , S β e -jπ / 2 , and calculate the positive sequence α signal S of the α β coordinate system α1 , positive sequence β signal S β1 , negative sequence α signal S α2 and negative sequence β signal S β2 , and the calculation formulas are:

[0030]

[0031] Then transform the four signals of the αβ coordinate system to the dq coordinate system to obtain the positive sequence d-axis signal S d1 , positive sequence q-axis signal S q1 , negative sequence d-axis signal S d2 and the negative sequence q-axis signal S q2 , its transformation formula is:

[0032]

[0033] Wherein, θ1′ is the positive sequence reference electrical angle of the previous control cycle, and θ2′ is the negative sequence reference electrical angle of the previous control cycle;

[0034] The inverter side current I abc , grid-side current I gabc and capacitor voltage V abc Substitute the three-phase signal S abc According to the above steps, 12 positive and negative sequence dq components are extracted.

[0035] Preferably, the positive sequence active power P in step 2 is e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 The calculation formulas are as follows:

[0036]

[0037] The calculation formulas for the positive sequence reference electrical angle θ1, the negative sequence reference electrical angle θ2 and the positive sequence reactive loop output E1 are respectively:

[0038]

[0039] Among them, D pis the active droop coefficient, J is the virtual inertia of the virtual synchronous generator, δ1 is the positive sequence power angle of the virtual synchronous generator terminal potential, δ2 is the negative sequence power angle of the virtual synchronous generator terminal potential, ω n is the rated angular velocity, ω1 is the positive sequence angular velocity of the virtual synchronous generator terminal potential, ω2 is the negative sequence angular velocity of the virtual synchronous generator terminal potential, V n is the rated terminal potential of the virtual synchronous generator, K p K is the proportional coefficient of the reactive loop PI controller, i is the integral coefficient of the reactive loop PI controller, and s is the Laplace operator.

[0040] Preferably, the positive sequence d-axis voltage drop ΔV in step 3 d1 , positive sequence q-axis voltage drop ΔV q1 , negative sequence d-axis voltage drop ΔV d2 and negative sequence q-axis voltage drop ΔV q2 The calculation formulas are:

[0041]

[0042] The positive sequence d-axis voltage loop command value V dref1 , positive sequence q axis voltage loop command value V qref1 , negative sequence d-axis voltage loop command value V dref2 and the negative sequence q-axis voltage loop command value V qref2 The calculation formulas are:

[0043]

[0044] The first positive sequence d-axis current loop command value I dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2 The calculation formulas are:

[0045]

[0046] Among them, K pv is the voltage regulator proportional coefficient, K iv is the voltage regulator integral coefficient.

[0047] Preferably, the current command value correction coefficient K in step 4 is lim The given process is as follows:

[0048] The first positive sequence d-axis current loop command value I dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value Idref2 and the first negative sequence q-axis current loop command value I qref2 Back-transform back to the αβ coordinate system to obtain the positive sequence α-axis current loop command value I αref1 , positive sequence β axis current loop command value I βref1 , negative sequence α axis current loop command value I αref2 and the negative sequence β axis current loop command value I βref2 , its transformation formula is:

[0049]

[0050] ξ1 is the first intermediate variable, ξ2 is the second intermediate variable, and their expressions are:

[0051]

[0052] Remember I m1 is the positive sequence current amplitude, I m2 is the negative sequence current amplitude, I ma is the current amplitude of phase a, I mb is the current amplitude of phase b, I mc is the c-phase current amplitude, then the maximum phase current amplitude I m for:

[0053]

[0054] The current command value correction coefficient K is given as follows lim :

[0055]

[0056] Among them, I lim is the given current limit value.

[0057] Preferably, in step 4, the modulated wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , after inverse Clarke transform and inverse Park transform, the positive sequence three-phase modulation wave U is obtained a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 The transformation formula is:

[0058]

[0059]

[0060] Where U α1 is the α-axis component of the positive sequence three-phase modulation wave, U β1 is the β-axis component of the positive sequence three-phase modulation wave, U α2 is the α-axis component of the negative-sequence three-phase modulation wave, U β2 It is the β-axis component of the negative-sequence three-phase modulation wave.

[0061] The beneficial effects of the present invention relative to the prior art are:

[0062] 1. Make the virtual synchronous generator conform to the operating characteristics of the traditional synchronous generator with symmetrical output voltage under asymmetrical faults.

[0063] 2. The electrical angle of the negative-sequence loop is generated by the negative-sequence power loop, which can eliminate the steady-state error of the negative-sequence control loop. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 2 is a schematic diagram of a grid-connected virtual synchronous generator in an embodiment of the present invention.

[0065] Figure 2 This is a block diagram of asymmetric fault ride-through control of a grid-type virtual synchronous generator based on positive and negative sequence separation in an embodiment of the present invention.

[0066] Figure 3 It is a simplified flow chart of the control strategy of the present invention.

[0067] Figure 4 It is a grid-connected voltage and current simulation waveform of a single-phase ground short circuit virtual synchronous generator based on positive and negative sequence separation. DETAILED DESCRIPTION

[0068] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0069] Figure 1 : This is a schematic diagram of a grid-connected virtual synchronous generator in an embodiment of the present invention. As can be seen from the figure, the circuit topology involved in this control strategy includes a DC side power supply, an inverter, a three-phase LC filter, a grid impedance, and a three-phase grid. The three-phase LC filter includes a three-phase filter inductor and a three-phase filter capacitor. The DC side power supply, inverter, filter inductor, grid impedance, and three-phase grid are connected in series in sequence, and the other end of the three-phase grid is grounded. One end of the filter capacitor is connected between the filter inductor and the grid impedance according to the phase sequence, and the other end is grounded.

[0070] exist Figure 1 In, V dc is the voltage of the DC power supply, 1 f is the three-phase filter inductor, C f is the three-phase filter capacitor, Z g is the grid impedance, Uga , U gb , U gc is the voltage of the three-phase grid.

[0071] Figure 2 This is a control block diagram of a grid-type virtual synchronous generator converter implemented in the present invention. Figure 3 This is a simplified flow chart of the control strategy of the present invention. Figure 2 and Figure 3 It can be seen that the present invention provides a control strategy for asymmetric fault ride-through of a meshed virtual synchronous generator. The meshed virtual synchronous generator is a voltage source inverter control scheme that simulates the operating characteristics of a synchronous generator, which includes a power loop consisting of an active loop and a reactive loop. The control strategy includes four steps. Figure 3 The following are named as positive and negative sequence component extraction, positive and negative sequence virtual synchronous generator control, positive and negative sequence virtual impedance and positive and negative sequence voltage loop, current limiter and positive and negative sequence current loop, specifically:

[0072] Step 1: Given the positive sequence active power command value P of the virtual synchronous generator ref1 , positive sequence reactive power command value Q ref1 , and let the negative sequence active power command value P ref2 =0;

[0073] Sampling the inverter side current I abc , grid-side current I gabc and capacitor voltage V abc The following 12 positive and negative sequence dq components are extracted by SOGI second-order generalized integrator, including: inverter side current positive sequence d-axis component I d1 , inverter side current positive sequence q-axis component I q1 , the negative sequence d-axis component of the inverter side current I d2 , the negative sequence q-axis component of the inverter side current I q2 , grid-side current positive sequence d-axis component I gd1 , grid-side current positive sequence q-axis component I gq1 , grid-side current negative sequence d-axis component I gd2 , grid-side current negative sequence q-axis component I gq2 、Capacitor voltage positive sequence d-axis component V d1 , capacitor voltage positive sequence q-axis component V q1 , capacitor voltage negative sequence d-axis component V d2 and the capacitor voltage negative sequence q-axis component V q2 .

[0074] In this embodiment, the steps of extracting 12 positive and negative sequence dq components by using a SOGI second-order generalized integrator are as follows:

[0075] Introducing three-phase signal S abc;

[0076] First, the three-phase signal S is transformed by Clarke abc Transformed into αβ coordinate system signal S a , S β , its transformation formula is:

[0077]

[0078] Then the αβ coordinate system signal S α , S β It is sent to the SOGI second-order generalized integrator to generate the orthogonal signal S α e -jπ / 2 , S β e -jπ / 2 , and calculate the positive sequence α signal S of the α β coordinate system α1 , positive sequence β signal S β1 , negative sequence α signal S α2 and negative sequence β signal S β2 , and the calculation formulas are:

[0079]

[0080] Then transform the four signals of the αβ coordinate system to the dq coordinate system to obtain the positive sequence d-axis signal S d1 , positive sequence q-axis signal S q1 , negative sequence d-axis signal S d2 and the negative sequence q-axis signal S q2 , its transformation formula is:

[0081]

[0082] Wherein, θ1′ is the positive sequence reference electrical angle of the previous control cycle, and θ2′ is the negative sequence reference electrical angle of the previous control cycle;

[0083] The inverter side current I abc , grid-side current I gabc and capacitor voltage V abc Substitute the three-phase signal S abc According to the above steps, 12 positive and negative sequence dq components are extracted.

[0084] Step 2: Calculate the positive sequence active power P according to the 12 dq axis components extracted in step 1. e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 ;

[0085] According to the positive sequence active power P e1 , positive sequence reactive power Q e1 and negative sequence active power Pe2 , the positive sequence reference electrical angle θ1, the negative sequence reference electrical angle θ2 and the positive sequence reactive loop output E1 are calculated by the virtual synchronous generator control algorithm, and the negative sequence reactive loop output E2 = 0 is given.

[0086] In this embodiment, the positive sequence active power P e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 The calculation formulas are as follows:

[0087]

[0088] The calculation formulas for the positive sequence reference electrical angle θ1, the negative sequence reference electrical angle θ2 and the positive sequence reactive loop output E1 are respectively:

[0089]

[0090] Among them, D p is the active droop coefficient, J is the virtual inertia of the virtual synchronous generator, δ1 is the positive sequence power angle of the virtual synchronous generator terminal potential, δ2 is the negative sequence power angle of the virtual synchronous generator terminal potential, ω n is the rated angular velocity, ω1 is the positive sequence angular velocity of the virtual synchronous generator terminal potential, ω2 is the negative sequence angular velocity of the virtual synchronous generator terminal potential, V n is the rated terminal potential of the virtual synchronous generator, K p K is the proportional coefficient of the reactive loop PI controller, i is the integral coefficient of the reactive loop PI controller, and s is the Laplace operator.

[0091] Step 3: Given a virtual resistor R v and virtual reactance X v , calculate the positive sequence d-axis voltage drop ΔV d1 , positive sequence q-axis voltage drop ΔV q1 , negative sequence d-axis voltage drop ΔV d2 and negative sequence q-axis voltage drop ΔV a2 Then, the positive sequence d-axis voltage loop command value V is calculated based on the four voltage drops. dref1 , positive sequence q axis voltage loop command value V qref1 , negative sequence d-axis voltage loop command value V dref2 and the negative sequence q-axis voltage loop command value V qref2 Then, the first positive sequence d-axis current loop command value I generated by the positive and negative sequence voltage loop is calculated based on the four voltage loop command values. dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2.

[0092] In this embodiment, the positive sequence d-axis voltage drop ΔV d1 , positive sequence q-axis voltage drop ΔV q1 , negative sequence d-axis voltage drop ΔV d2 and negative sequence q-axis voltage drop ΔV q2 The calculation formulas are:

[0093]

[0094] The positive sequence d-axis voltage loop command value V dref1 , positive sequence q axis voltage loop command value V qref1 , negative sequence d-axis voltage loop command value V dref2 and the negative sequence q-axis voltage loop command value V qref2 The calculation formulas are:

[0095]

[0096] The first positive sequence d-axis current loop command value I dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2 The calculation formulas are:

[0097]

[0098] Among them, K pv is the voltage regulator proportional coefficient, K iv is the voltage regulator integral coefficient.

[0099] Step 4: Correct the coefficient K according to the current command value lim , calculate the second positive sequence d-axis current loop command value I after passing through the positive and negative sequence current loop and current limiting dref1 ', the second positive sequence q-axis current loop command value I qref1 ', the second negative sequence d-axis current loop command value I dref2 ′ and the second negative sequence q-axis current loop command value I qref2 ′, and its calculation formulas are:

[0100]

[0101] Perform positive and negative sequence current loop control and output modulation wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , whose expression is:

[0102]

[0103] Among them, K pi is the current regulator proportional coefficient, K ii is the current regulator integral coefficient;

[0104] The modulated wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , after inverse Clarke transform and inverse Park transform, the positive sequence three-phase modulation wave U is obtained a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 ;

[0105] According to the positive sequence three-phase modulation wave U a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 Get the three-phase modulation wave U a , U b , U c , whose expression is:

[0106]

[0107] In this embodiment, P ref1 =1725000W,Q ref1 =0W, P ref2 =0W, D p =350W×s / rad,J=0.035 / Kg×m 2 , V n =563V,ω n =100π,K p =0.000026, K i =0.00001, K pv =20,K iv =10,K pi =0.2, K ii =0.4, R v =0.005Ω, X v =0.003Ω, I lim =3061A.

[0108] In this embodiment, the current command value correction coefficient K lim The given process is as follows:

[0109] The first positive sequence d-axis current loop command value I dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2 Back-transform back to the αβ coordinate system to obtain the positive sequence α-axis current loop command value I αref1 , positive sequence β axis current loop command value I βref1 , negative sequence α axis current loop command value I αref2 and the negative sequence β axis current loop command value I βref2 , and its transformation formula is:

[0110]

[0111] In the formula, ξ1 is the first intermediate variable, ξ2 is the second intermediate variable, and its expression is:

[0112]

[0113] Remember I m1 is the positive sequence current amplitude, I m2 is the negative sequence current amplitude, I ma is the current amplitude of phase a, I mb is the current amplitude of phase b, I mc is the c-phase current amplitude, then the maximum phase current amplitude I m for:

[0114]

[0115] The current command value correction coefficient K is given as follows lim :

[0116]

[0117] Among them, I lim is the given current limit value.

[0118] In this embodiment, the modulated wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , after inverse Clarke transform and inverse Park transform, the positive sequence three-phase modulation wave U is obtained a1 , U b1 , U c1 And negative sequence three-phase modulation wave Ua2 , U b2 , U c2 The transformation formula is:

[0119]

[0120]

[0121] Where U α1 is the α-axis component of the positive sequence three-phase modulation wave, U β1 is the β-axis component of the positive sequence three-phase modulation wave, U α2 is the α-axis component of the negative-sequence three-phase modulation wave, U β2 It is the β-axis component of the negative-sequence three-phase modulation wave.

[0122] Figure 4 In the embodiment of the present invention, the single-phase voltage of the power grid drops to 0.5U n The inverter output voltage and current waveform after Figure 4 It can be seen that when the voltage output by the grid-forming virtual synchronous generator is approximately symmetrical during the period from 1.3s to 1.73s, and the output current does not exceed 1.5 times the rated limit, it shows that the proposed strategy can enable the virtual synchronous generator to meet the operating characteristics of the traditional synchronous generator with symmetrical output voltage under asymmetric faults, and the electrical angle of the negative-sequence loop is generated by the negative-sequence power loop, which can eliminate the steady-state error of the negative-sequence control loop.

Claims

1. A grid-type virtual synchronous generator asymmetric fault ride-through control strategy, the circuit topology involved in the control strategy includes a DC side power supply, an inverter, a three-phase LC filter, a grid impedance, and a three-phase grid. The three-phase LC filter includes a three-phase filter inductor and a three-phase filter capacitor. The DC side power supply, inverter, filter inductor, grid impedance, and three-phase grid are connected in series in sequence, and the other end of the three-phase grid is grounded. One end of the filter capacitor is connected between the filter inductor and the grid impedance according to phase sequence, and the other end is grounded. It is characterized by: The control strategy includes the following steps: Step 1: Given the positive sequence active power command value P of the virtual synchronous generator ref1 , positive sequence reactive power command value Q ref1 , and let the negative sequence active power command value P ref2 =0; Sampling the inverter side current I abc , grid-side current I gabc and capacitor voltage V abc The following 12 positive and negative sequence dq components are extracted by SOGI second-order generalized integrator, including: inverter side current positive sequence d-axis component I d1 , inverter side current positive sequence q-axis component I q1 , the negative sequence d-axis component of the inverter side current I d2 , the negative sequence q-axis component of the inverter side current I q2 , grid-side current positive sequence d-axis component I gd1 , grid-side current positive sequence q-axis component I gq1 , grid-side current negative sequence d-axis component I gd2 , grid-side current negative sequence q-axis component I gq2 、Capacitor voltage positive sequence d-axis component V d1 , capacitor voltage positive sequence q-axis component V q1 , capacitor voltage negative sequence d-axis component V d2 and the capacitor voltage negative sequence q-axis component V q2 ; Step 2: Calculate the positive sequence active power P according to the 12 dq axis components extracted in step 1. e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 ; According to the positive sequence active power P e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 , calculate the positive sequence reference electrical angle θ1, the negative sequence reference electrical angle θ2 and the positive sequence reactive loop output E1 through the virtual synchronous generator control algorithm, and set the negative sequence reactive loop output E2 = 0; Step 3: Given a virtual resistor R v and virtual reactance X v , calculate the positive sequence d-axis voltage drop ΔV d1 , positive sequence q-axis voltage drop ΔV q1 , negative sequence d-axis voltage drop ΔV d2 and negative sequence q-axis voltage drop ΔV q2 Then, the positive sequence d-axis voltage loop command value V is calculated based on the four voltage drops. dref1 , positive sequence q axis voltage loop command value V qref1 , negative sequence d-axis voltage loop command value V dref2 and the negative sequence q-axis voltage loop command value V qref2 Then, the first positive sequence d-axis current loop command value I generated by the positive and negative sequence voltage loop is calculated based on the four voltage loop command values. dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2 ; Step 4: Correct the coefficient K according to the current command value lim , calculate the second positive sequence d-axis current loop command value I after passing through the positive and negative sequence current loop and current limiting dref1 ', the second positive sequence q-axis current loop command value I qref1 ', the second negative sequence d-axis current loop command value I dref2 ′ and the second negative sequence q-axis current loop command value I qref2 ′, and its calculation formulas are: Perform positive and negative sequence current loop control and output modulation wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , whose expression is: Among them, K pi is the current regulator proportional coefficient, K ii is the current regulator integral coefficient; The modulated wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , after inverse Clarke transform and inverse Park transform, the positive sequence three-phase modulation wave U is obtained a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 ; According to the positive sequence three-phase modulation wave U a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 Get the three-phase modulation wave U a , U b , U c , whose expression is:

2. A grid-type virtual synchronous generator asymmetric fault ride-through control strategy according to claim 1, characterized in that: The steps for extracting 12 positive and negative sequence dq components by using the SOGI second-order generalized integrator in step 1 are as follows: Introducing three-phase signal S abc ; First, the three-phase signal S is transformed by Clarke abc Transformed into αβ coordinate system signal S α , S β , and its transformation formula is: Secondly, the αβ coordinate system signal S α , S β It is sent to the SOGI second-order generalized integrator to generate the orthogonal signal S α e -jπ / 2 , S β e -jπ / 2 , and calculate the positive sequence α signal S in the αβ coordinate system α1 , positive sequence β signal S β1 , negative sequence α signal S α2 and negative sequence β signal S β2 , and the calculation formulas are: Then transform the four signals of the αβ coordinate system to the dq coordinate system to obtain the positive sequence d-axis signal S d1 , positive sequence q-axis signal S q1 , negative sequence d-axis signal S d2 and the negative sequence q-axis signal S q2 , its transformation formula is: Wherein, θ1′ is the positive sequence reference electrical angle of the previous control cycle, and θ2′ is the negative sequence reference electrical angle of the previous control cycle; The inverter side current I abc , grid-side current I gabc and capacitor voltage V abc Substitute the three-phase signal S abc According to the above steps, 12 positive and negative sequence dq components are extracted.

3. A grid-type virtual synchronous generator asymmetric fault ride-through control strategy according to claim 1, characterized in that: The positive sequence active power P in step 2 e1 , positive sequence reactive power Q e1 and negative sequence active power P e2 The calculation formulas are as follows: The calculation formulas for the positive sequence reference electrical angle θ1, the negative sequence reference electrical angle θ2 and the positive sequence reactive loop output E1 are respectively: Among them, D p is the active droop coefficient, J is the virtual inertia of the virtual synchronous generator, δ1 is the positive sequence power angle of the virtual synchronous generator terminal potential, δ2 is the negative sequence power angle of the virtual synchronous generator terminal potential, ω n is the rated angular velocity, ω1 is the positive sequence angular velocity of the virtual synchronous generator terminal potential, ω2 is the negative sequence angular velocity of the virtual synchronous generator terminal potential, V n is the rated terminal potential of the virtual synchronous generator, K p K is the proportional coefficient of the reactive loop PI controller, i is the integral coefficient of the reactive loop PI controller, and s is the Laplace operator.

4. A grid-type virtual synchronous generator asymmetric fault ride-through control strategy according to claim 1, characterized in that: The positive sequence d-axis voltage drop ΔV in step 3 d1 , positive sequence q-axis voltage drop ΔV q1 , negative sequence d-axis voltage drop ΔV d2 and negative sequence q-axis voltage drop ΔV q2 The calculation formulas are: The positive sequence d-axis voltage loop command value V dref1 , positive sequence q axis voltage loop command value V qref1 , negative sequence d-axis voltage loop command value V dref2 and the negative sequence q-axis voltage loop command value V qref2 The calculation formulas are: The first positive sequence d-axis current loop command value I dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2 The calculation formulas are: Among them, K pv is the voltage regulator proportional coefficient, K iv is the voltage regulator integral coefficient.

5. The asymmetric fault ride-through control strategy for a grid-type virtual synchronous generator according to claim 1 is characterized in that: The current command value correction coefficient K in step 4 lim The given process is as follows: The first positive sequence d-axis current loop command value I dref1 , the first positive sequence q-axis current loop command value I qref1 , the first negative sequence d-axis current loop command value I dref2 and the first negative sequence q-axis current loop command value I qref2 Back-transform back to the αβ coordinate system to obtain the positive sequence α axis current loop command value I αref1 , positive sequence β axis current loop command value I βref1 , negative sequence α axis current loop command value I αref2 and the negative sequence β axis current loop command value I βref2 , and its transformation formula is: ξ1 is the first intermediate variable, ξ2 is the second intermediate variable, and their expressions are: Remember I m1 is the positive sequence current amplitude, I m2 is the negative sequence current amplitude, I ma is the current amplitude of phase a, I mb is the current amplitude of phase b, I mc is the c-phase current amplitude, then the maximum phase current amplitude I m for: The current command value correction coefficient K is given as follows lim : Among them, I lim is the given current limit value.

6. A grid-type virtual synchronous generator asymmetric fault ride-through control strategy according to claim 1, characterized in that: As described in step 4, the modulated wave U abc The positive sequence d-axis component U dref1 , positive sequence q-axis component U qref1 , negative sequence d-axis component U dref2 and the negative sequence q-axis component U qref2 , after inverse Clarke transform and inverse Park transform, the positive sequence three-phase modulation wave U is obtained a1 , U b1 , U c1 And negative sequence three-phase modulation wave U a2 , U b2 , U c2 The transformation formula is: Where U α1 is the α-axis component of the positive sequence three-phase modulation wave, U β1 is the β-axis component of the positive sequence three-phase modulation wave, U α2 is the α-axis component of the negative-sequence three-phase modulation wave, U β2 It is the β-axis component of the negative-sequence three-phase modulation wave.

Images