CN103311949B - The low voltage traversing control method of high-power photovoltaic inverter - Google Patents

Abstract

A kind of low voltage traversing control method of high-power photovoltaic inverter, first, based on the high-power photovoltaic inverter structure of LCL filter, consider the low voltage crossing situation of grid voltage sags especially under imbalance falls condition, to suppress negative-sequence current for control objectives, calculate forward-order current and negative-sequence current reference set-point, then pass through the idle of control algolithm compensation filter capacitive absorption, grid-connected to ensure the High Power Factor under different capacity grade is run; Secondly, control based on positive-negative sequence double-current ring, design and add filter capacitor current feed-forward signal in modulation signal, by regulating feed-forward coefficients, the active damping realized under unbalance control controls; Finally, design disturbance quantity feedback coefficient, suppresses transient current to impact.The method effectively can suppress the rush of current of grid voltage sags moment, and ensure that the three-phase equilibrium of electric current before and after grid voltage sags exports, and control effects meets the quality of power supply requirement of electrical network to photovoltaic inverter grid-connected.

Description

The low voltage traversing control method of high-power photovoltaic inverter

Technical field

The present invention relates to a kind of low voltage traversing control method of high-power photovoltaic inverter, being uninterruptedly incorporated into the power networks for keeping photovoltaic DC-to-AC converter during electric network fault.

Background technology

Along with a large amount of new forms of energy access electric power system, the excision of high-power photovoltaic energy moment will cause the disappearance of electric network active, and this causes very important impact to the stable recovery of electric power system.For the safety and stablization of electric power system under maintenance photovoltaic energy Thief zone are run, the new big-and-middle-sized photovoltaic DC-to-AC converter of grid integration rule request should possess low voltage crossing (Low Voltage Ride Through, LVRT) ability, namely keeps photovoltaic DC-to-AC converter to be uninterruptedly incorporated into the power networks during electric network fault.When adopting traditional two close cycles vector control strategy, grid voltage sags can cause the impact of grid-connected current instantaneously, hardware protection may be caused to trip, even burn power device; Especially will the height of grid-connected current be caused uneven when imbalance is fallen, thus cause the power of feed-in electrical network to vibrate, affect the safe operation of photovoltaic DC-to-AC converter and electrical network.For protecting the safety of photovoltaic plant, safeguarding the stable of electric power system, need study the grid-connected control method of photovoltaic DC-to-AC converter during low voltage crossing.

The service conditions that normal and three-phase equilibrium is fallen according to line voltage, (Zheng Fei, Zhang Junjun, Ding Mingchang. based on the photovoltaic generating system low voltage crossing modeling and control strategy of RTDS. Automation of Electric Systems, 2012,36 (22): 19-24) strategy adopting conventional control and LVRT to control to switch mutually realizes low voltage crossing, but does not fall imbalance and analyze, and is difficult to realize photovoltaic generating system and falls period low voltage crossing requirement at line voltage in different faults.And for the control strategy research that imbalance is fallen, classical control is, by symmetrical component method, amount of unbalance is carried out positive and negative sequence decomposition, and the power equation established under positive and negative sequence symmetric condition, according to the control objectives chosen, calculate required current-order, the DAZ gene to instruction current is realized, to realize the control under uneven condition by controller.Owing to there is power fluctuation in grid-connected reactor, can not ignore completely, therefore (Liao Yong, Zhuan Kai, Yao Jun. the control strategy of total power wind-electricity integration current transformer during unbalanced source voltage. electric power network technique .2012,36 (1): 72-78) consider the power fluctuation on reactor and set up new voltage equation, introducing side currents feedforward simultaneously realizes the stability contorting under uneven condition, but experiment is only for slight uneven condition, can the low voltage crossing that meet when imbalance is fallen be required to need checking; In addition, in experiment, adopt inductance filter, and normal filter effect and the better LCL filter of economic performance of adopting under high-power condition, therefore also need to analyze the control method based on LCL filter.At present, the control mode studying LCL filter is studied based on line voltage symmetry mostly, the stability of system cloud gray model of must considering that in low voltage crossing process grid voltage sags is especially asymmetric when falling, therefore existing method is difficult to solve the problem that line voltage is asymmetric when falling, in sum, existing control algolithm cannot meet high-power photovoltaic inverter grid-connected system (grid connection topology based on LCL filter) the low voltage crossing control overflow under different grid voltage sags condition.

Summary of the invention

Technology of the present invention is dealt with problems: the problem overcoming the transient current impact that existing high-power photovoltaic inverter can not effectively suppress line voltage to bring when falling, propose a kind of low voltage crossing voltage control method, achieve the stable operation of system before and after grid voltage sags, and meet the reactive power support requirement in grid-connected rule.

Technical solution of the present invention is: a kind of low voltage traversing control method of high-power photovoltaic inverter, the steps include:

1. consider that line voltage is falling, especially under imbalance falls condition, the low voltage crossing situation of photovoltaic DC-to-AC converter, calculate forward-order current and negative-sequence current reference set-point, simultaneously for ensureing that photovoltaic DC-to-AC converter has high power factor when grid-connected, choose the reactive power that inverter compensation filter electric capacity absorbs; Wherein, the calculation expression of forward-order current and negative-sequence current reference set-point is:

$\left\{\begin{array}{c}{i}_{\mathrm{inv}\_d}^{p*}=\frac{2e_d^p{P}_0+e_q^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_q}^{p*}=\frac{2e_q^p{P}_0-e_d^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_d}^{n*}=0\\ i_{\mathrm{inv}\_q}^{n*}=0\end{array}\right.$

P in formula ₀, Q ₀be respectively and export meritorious, idle DC component; be respectively the current reference set-point of positive sequence dq axle and negative phase-sequence dq axle; be respectively line voltage positive sequence dq axle component.

The calculation expression choosing the reactive power that inverter compensation filter electric capacity absorbs is:

$\mathrm{\ΔQ}=\mathrm{\ω}{C}_f[{\left(u_{\mathrm{cd}}^p\right)}^2+{\left(u_{\mathrm{cq}}^p\right)}^2+{\left(u_{\mathrm{cq}}^n\right)}^2+{\left(u_{\mathrm{cd}}^n\right)}^2]$

In formula, Δ Q is the idle of compensation; ω is the mains frequency detected; C _ffor the filter capacitor in LCL filter; be respectively the positive-negative sequence dq axle component of filter capacitor.

2. based on positive-negative sequence double-current ring control strategy, design and add filter capacitor current feed-forward in modulation signal, by regulating current feed-forward coefficient k _c, the damping realized under uneven condition controls;

3. for photovoltaic inverter grid-connected system, transient process during Voltage Drop is considered, design disturbance electric voltage feed forward coefficient k _e, suppress current transient during grid voltage sags to be impacted;

4. the current feed-forward coefficient k that 2. forward-order current 1. calculated according to step and negative-sequence current set-point, step set _cand the disturbance electric voltage feed forward coefficient k that 3. step designs _e, carry out positive-negative sequence double-current ring and control, week is to realize the low voltage crossing of high-power photovoltaic inverter.

Described step 2. in k _cselection range is 0.2 ~ 0.7, and wherein said active damping controls, and being a kind of unbalance control algorithm based on positive-negative sequence double-current ring, by adding filter capacitor current feed-forward signal in SPWM modulating wave, comparatively generating final PWM waveform with carrier wave ratio.

Described step 3. in disturbance electric voltage feed forward coefficient k _eexpression formula be:

$k_e=\frac{1}{G_{\mathrm{inv}}}$

G _invfor the equivalent transfer function of three-phase photovoltaic inverter.

Principle of the present invention is:

The low voltage traversing control method of high-power photovoltaic inverter of the present invention is applicable to non-isolation type stage photovoltaic single grid connected structure, and its topological structure is made up of photovoltaic array, combining inverter, LCL filter and control system, as shown in Figure 2, and L in figure _invfor inverter side inductance value; L _gfor grid side inductance value; C _ffor filter capacitor.Founding mathematical models:

$\begin{array}{c}{u}_{\mathrm{inv}\_x}=L_{\mathrm{inv}}\frac{{\mathrm{di}}_{\mathrm{inv}\_x}}{\mathrm{dt}}+u_{c\_x}\\ i_{\mathrm{inv}\_x}=i_{g\_x}+i_{c\_x}\\ u_{c\_x}=L_g\frac{{\mathrm{di}}_{g\_x}}{\mathrm{dt}}+e_x\end{array}---\left(1\right)$

In formula: x=a, b, c, wherein a, b, c represent three-phase; u _{inv_x}, i _{inv_x}, e _x, i _{g_x}, u _{c_x}, i _{c_x}represent the voltage and current on a certain phase voltage of inverter outlet side and electric current, line voltage and electric current, filter capacitor respectively.

Consider that complicated imbalance falls low voltage crossing situation under condition, whole method control block diagram as shown in Figure 3.Wherein, by phase-locked loop and positive-negative sequence separation module, input variable is processed, obtain each variable required when controlling; Control loop is made up of outer voltage and current inner loop, when electrical network is normal, controls by outer voltage the maximum power point tracking realizing photovoltaic battery panel; When Voltage Drop, excise outer voltage in real time according to testing conditions and adopt current inner loop to control; When after power system restoration, cut-in voltage outer shroud, the reference of Voltage loop is given returns to the value before fault with ramp system.Control method is mainly divided into three parts, and Part I is the calculating of current instruction value, and Part II is the implementation of positive-negative sequence double-current ring active damping, and filter capacitor current feed-forward coefficient k _cdetermination, Part III is disturbance electric voltage feed forward coefficient k _edetermination.

For Part I, from Fig. 2 we can obtain inverter export power can be expressed as:

$S=\frac{3}{2}(u_{\mathrm{inv}\_\mathrm{dq}}^p{e}^{\mathrm{j\ωt}}+u_{\mathrm{inv}\_\mathrm{dq}}^n{e}^{-\mathrm{j\ωt}})\left(\stackrel{\‾}{i_{\mathrm{inv}\_\mathrm{dq}}^p{e}^{\mathrm{j\ωt}}+i_{\mathrm{inv}\_\mathrm{dq}}^n{e}^{-\mathrm{j\ωt}}}\right)---\left(2\right)$

Active power of output and reactive power can be expressed as:

P＝(P ₀+P _c2cos2ωt+P _s2sin2ωt) (3)

Q＝(Q ₀+Q _c2cos2ωt+Q _s2sin2ωt) (4)

In formula: P ₀, Q ₀be respectively and export meritorious, idle DC component; be respectively inverter outlet side voltage positive-negative sequence d axle, q axle component; be respectively inverter outlet side electric current positive-negative sequence d axle, q axle component; P _c2, P _s2, Q _c2, Q _s2for exporting meritorious, idle wave component; ω is line voltage angular frequency.Real component and idle component are expanded into:

$\left[\begin{array}{c}{P}_0\\ P_{c2}\\ Q_{s2}\\ Q_0\\ Q_{c2}\\ Q_{s2}\end{array}\right]=\frac{3}{2}\left[\begin{array}{cccc}{u}_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^p& u_{\mathrm{inv}\_d}^n& u_{\mathrm{inv}\_q}^n\\ u_{\mathrm{inv}\_d}^n& u_{\mathrm{inv}\_q}^n& u_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^p\\ u_{\mathrm{inv}\_q}^n& -u_{\mathrm{inv}\_d}^n& -u_{\mathrm{inv}\_q}^p& u_{\mathrm{inv}\_d}^p\\ u_{\mathrm{inv}\_q}^p& -u_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^n& -u_{\mathrm{inv}\_d}^n\\ u_{\mathrm{inv}\_q}^n& {-u}_{\mathrm{inv}\_d}^n& u_{\mathrm{inv}\_q}^p& -u_{\mathrm{inv}\_d}^p\\ {-u}_{\mathrm{inv}\_d}^n& {-u}_{\mathrm{inv}\_q}^n& u_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^p\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{inv}\_d}^p\\ i_{\mathrm{inv}\_q}^p\\ i_{\mathrm{inv}\_d}^n\\ i_{\mathrm{inv}\_q}^n\end{array}\right]---\left(5\right)$

In order to utilize this known measuring amount of line voltage, obtaining accurate instruction current, need convert formula (5).Carry out coordinate transform by formula (1), and carry out positive-negative sequence and be separated the Mathematical Modeling that can obtain under positive-negative sequence component and can be expressed as:

$\begin{array}{c}\left\{\begin{array}{c}{L}_{\mathrm{inv}}\frac{{\mathrm{di}}_{\mathrm{inv}\_\mathrm{dq}}^p}{\mathrm{dt}}=u_{\mathrm{inv}\_\mathrm{dq}}^p-u_{c\_\mathrm{dq}}^p\±\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_\mathrm{qd}}^p\\ L_g\frac{{\mathrm{di}}_{g\_\mathrm{dq}}^p}{\mathrm{dt}}=u_{c\_\mathrm{dq}}^p-e_{\mathrm{dq}}^p\±\mathrm{\ω}{L}_g{i}_{g\_\mathrm{qd}}^p\\ C_f\frac{{\mathrm{du}}_{c\_\mathrm{dq}}^p}{\mathrm{dt}}=i_{\mathrm{inv}\_\mathrm{dq}}^p-i_{g\_\mathrm{dq}}^p\±\mathrm{\ω}{C}_f{u}_{c\_\mathrm{qd}}^p\end{array}\right.\\ \left\{\begin{array}{c}{L}_{\mathrm{inv}}\frac{{\mathrm{di}}_{\mathrm{inv}\_\mathrm{dq}}^n}{\mathrm{dt}}=u_{\mathrm{inv}\_\mathrm{dq}}^n-u_{c\_\mathrm{dq}}^n\stackrel{\‾}{+}\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_\mathrm{qd}}^n\\ L_g\frac{{\mathrm{di}}_{g\_\mathrm{dq}}^n}{\mathrm{dt}}=u_{c\_\mathrm{dq}}^n-e_{\mathrm{dq}}^n\stackrel{\‾}{+}\mathrm{\ω}{L}_g{i}_{g\_\mathrm{qd}}^n\\ C_f\frac{{\mathrm{du}}_{c\_\mathrm{dq}}^n}{\mathrm{dt}}=i_{\mathrm{inv}\_\mathrm{dq}}^n-i_{g\_\mathrm{dq}}^n\stackrel{\‾}{+}\mathrm{\ω}{C}_f{u}_{c\_\mathrm{qd}}^n\end{array}\right.\end{array}---\left(6\right)$

Can be obtained by formula (6), under systematic steady state condition, the pass between inverter side voltage and line voltage is:

$\begin{array}{c}{u}_{\mathrm{inv}\_d}^p=e_d^p-\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p\\ u_{\mathrm{inv}\_q}^p=e_q^p+\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p\\ u_{\mathrm{inv}\_d}^n=e_d^n+\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n\\ u_{\mathrm{inv}\_q}^n=e_q^n-\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n\end{array}---\left(7\right)$

Bring formula (7) into formula (5) can obtain

$\left[\begin{array}{c}{P}_0\\ Q_0\\ P_{c2}\\ P_{s2}\end{array}\right]=\frac{3}{2}\left[\begin{array}{cccc}{e}_d^p& e_q^p& e_d^n& e_q^n\\ e_q^p& -e_d^p& e_q^n& -e_d^n\\ e_d^n& e_q^n& e_d^p& e_q^p\\ e_q^n& -e_d^n& -e_q^p& e_d^p\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{inv}\_d}^p\\ i_{\mathrm{inv}\_q}^p\\ i_{\mathrm{inv}\_d}^n\\ i_{\mathrm{inv}\_q}^n\end{array}\right]---\left(8\right)$

$\frac{3}{2}\left[\begin{array}{cccc}-{\mathrm{\ω}}_{\mathrm{inv}}{\mathrm{Li}}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p& L_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p& \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n\\ \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p& \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n\\ \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p& \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p\\ -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{inv}\_d}^p\\ i_{\mathrm{inv}\_q}^p\\ i_{\mathrm{inv}\_d}^n\\ i_{\mathrm{inv}\_q}^n\end{array}\right]$

Formula (8) is the relation between known measuring amount and inverter output power.For eliminating negative-sequence current namely with in formula (8) for target, when considering Operation at full power, under the operational factor of design, the ratio of formula (8) right side shared by coupling terms is very little, and calculate for simplifying, ignore this impact, therefore current reference set-point is

$\left\{\begin{array}{c}{i}_{\mathrm{inv}\_d}^{p*}=\frac{2e_d^p{P}_0+e_q^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_q}^{p*}=\frac{2e_q^p{P}_0-e_d^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_d}^{n*}=0\\ i_{\mathrm{inv}\_q}^{n*}=0\end{array}\right.---\left(9\right)$

In formula (9), for the current instruction value of inverter outlet side, under low-power grid-connected conditions, the impact of filter on grid-connected power factor be can not ignore.For guaranteeing that photovoltaic plant keeps higher power factor when being incorporated into the power networks, need the reactive power that compensation filter electric capacity absorbs, absorbing reactive power Δ Q can choose:

$\mathrm{\ΔQ}=\mathrm{\ω}{C}_f[{\left(u_{\mathrm{cd}}^p\right)}^2+{\left(u_{\mathrm{cq}}^p\right)}^2+{\left(u_{\mathrm{cq}}^n\right)}^2+{\left(u_{\mathrm{cd}}^n\right)}^2]---\left(10\right)$

When steady operational status, make Q ₀=Δ Q, brings formula (7) into and can obtain inner ring given value of current value when unity power factor is incorporated into the power networks.During low voltage crossing, according to the Voltage Drop degree of depth detected, real time modifying Q ₀size export to regulate reactive power.For ensureing that grid-connected current is no more than safety threshold value, meritorious, the reactive current component size of output should meet

$i_{\mathrm{inv}\_d}^{p*}\≤\sqrt{{\left(i_{\max }\right)}^2-{\left(i_{\mathrm{inv}\_q}^{p*}\right)}^2}---\left(11\right)$

In formula (11), i _maxrepresent the maximum limit tentering value of output current.Stability contorting after Voltage Drop and reactive power support by dynamically changing d axle, q shaft current command value realizes, and chooses suitable controller parameter to ensure the stability of system.

For Part II, under high-power condition, LCL filter can significantly reduce filter inductance design load, reduces volume and the loss of system, but there is resonance problems.Under the prerequisite not affecting photovoltaic inverter grid-connected efficiency, the normal active damping that adopts is to suppress resonance.Under line voltage equilibrium condition, active damping current loop control block diagram as shown in Figure 4.Wherein, G _invfor the Mathematical Modeling of inverter, G _is () is current controller transfer function, 3s/2r, 2r/3s denotation coordination converts, and dotted box portion is the equivalent mathematical model of LCL filter.But owing to adopting positive-negative sequence double-current ring control mode under uneven condition, voltage, electric current are all separated into positive sequence component and negative sequence component, and filter capacitor current i _cbe mainly high fdrequency component, conventional positive-negative sequence separate mode all cannot obtain positive sequence under high frequency and negative sequence component, therefore adopts the active damping control mode under equilibrium condition, cannot realize damping and control under uneven condition, and even influential system is stable.Based on positive-negative sequence double-current ring control strategy, the current loop control block diagram that the present invention adopts as shown in Figure 5.In Figure 5, G _invfor the Mathematical Modeling of inverter, G _is () is current controller transfer function, 3s/2r, 2r/3s denotation coordination converts, and dotted line frame 1 is the control block diagram of unbalance control algorithm, and dotted line frame 2 is the equivalent mathematical model of LCL.Feed-forward mode shown in Fig. 5, effectively prevent feedforward amount i _ccarry out positive-negative sequence separation, and feed-forward coefficients k _cadjustment, the not transfer function of active damping algorithm in effect diagram 4, the resonance that can effectively suppress LCL filter to exist.

For Part III, in Voltage Drop moment, be incorporated into the power networks and the rush of current phenomenon in low voltage crossing running status handoff procedure to stable, must be controlled to ensure that photovoltaic DC-to-AC converter can enter safely the traffic coverage after falling.Loop shown in Fig. 5 is simplified, as shown in Figure 6, and obtains the closed loop transfer function, of system:

$I_g=\frac{G_i\left(s\right)G_1\left(s\right)G_2\left(s\right)}{1+G_i\left(s\right)G_1\left(s\right)G_2\left(s\right)}{I}_{\mathrm{inv}}^{*}-\frac{G_2\left(s\right)(k_e{G}_1\left(s\right)-1)}{1+G_i\left(s\right)G_1\left(s\right)G_2\left(s\right)}E---\left(10\right)$

In formula, k _efor disturbance electric voltage feed forward coefficient.

$G_1\left(s\right)=\frac{G_{\mathrm{inv}}}{L_{\mathrm{inv}}{C}_f{s}^2+(G_{\mathrm{inv}}{G}_i\left(s\right)-k_c{G}_{\mathrm{inv}})C_{f}s+1}---\left(11\right)$

$G_2\left(s\right)=\frac{L_{\mathrm{inv}}{C}_f{s}^2+(G_{\mathrm{inv}}{G}_i\left(s\right)-k_c{G}_{\mathrm{inv}})C_{f}s+1}{L_{\mathrm{inv}}{C}_f{L}_g{s}^3+(G_{\mathrm{inv}}{G}_i\left(s\right)-k_c{G}_{\mathrm{inv}})L_g{C}_f{s}^2+s(L+L_g)}---\left(12\right)$

From formula (12), control output variable I _gwith input variable I* _invall contact is there is with line voltage E.E is considered as disturbance term, and Voltage Drop moment, grid-connected current is by proportional increase.Therefore, in order to eliminate the impact of disturbance quantity on grid-connected current, introduce voltage feed-forward control, as shown in Fig. 6 dotted portion.

If eliminate disturbance term completely, can obtain

$k_e=\frac{1}{G_1\left(s\right)}=\frac{1}{G_{\mathrm{inv}}}(L_{\mathrm{inv}}{C}_f{s}^2+(G_{\mathrm{inv}}{G}_i\left(s\right)-k_c{G}_{\mathrm{inv}})C_{f}s+1)---\left(13\right)$

The disturbance electric voltage feed forward coefficient k of employing formula (13) _ecan realize the suppression completely to disturbance quantity E, namely the distortion of E can not affect I _goutput.There is differential term and second-order differential item in factor (13), wherein, differential term and second-order differential item have inhibitory action to high fdrequency component, and proportional then affects the effect of low frequency component.Fall for line voltage e, especially when e falls instantaneously, the existence of micro component will cause feed-forward signal to be tending towards infinitely great, make system unstable.In grid voltage sags moment, in order to suppress grid-connected current transient impact, the safe operation of protection grid-connection device, adapts to LVRT requirement, simplifies, ignore the differential term in feed-forward coefficients and second-order differential item, can obtain formula (13)

$k_e=\frac{1}{G_{\mathrm{inv}}}---\left(14\right)$

Proportional component is reduced to by feed-forward coefficients.Meanwhile, in acquisition channel, add low pass filter, to eliminate the high frequency content that exists in feed-forward signal e to the impact of grid-connected output current wave, the transient current caused grid voltage sags impacts and plays good inhibitory action by the introducing of feedforward amount.

The method of the invention, when practical application, when Voltage Drop being detected, falls flag bit by enable, and excision is realized the outer voltage of maximum power point tracking by control strategy, takes current inner loop to control.Reduce meritorious given simultaneously, and on-the-fly modify idle given, qualifications such as formula (11), with the idle output during realizing low voltage crossing.Regulated by controling parameters, choose optimum parameter to ensure stable state and the dynamic property of current loop, thus realize the stable cutting-in control before and after low voltage crossing.

The present invention's advantage is compared with prior art:

Low voltage traversing control method for high-power photovoltaic inverter of the present invention, for the high-power photovoltaic inverter based on LCL filter, consider and occur that serious imbalance falls situation at dissimilar grid voltage sags especially electrical network, by introducing negative-sequence current control ring, and the current instruction value calculating simplification is to suppress negative-sequence current, thus realizes grid-connected current three-phase equilibrium and export; For positive-negative sequence double-current control mode, the filter capacitor current feed-forward mode under the adaptation unbalance control of design ensure that the stable of grid-connected system; In grid voltage sags moment; the voltage feed-forward control adopted can effectively suppress transient current to impact; thus protection inverter safe transition is to low voltage crossing running status; in conjunction with positive-negative sequence component extraction fast and accurate mains voltage signal synchronous; the real-time of signal acquisition in controlling can be guaranteed; and the output of dynamic adjustments reactive current according to demand; control effects ensure that the stable operation of system before and after Voltage Drop, and meets the low voltage crossing requirement of electric power system for photovoltaic combining inverter.

Accompanying drawing explanation

Fig. 1 is high-power photovoltaic inverter low voltage traversing control method flow chart of the present invention.

Fig. 2 is the topology diagram of stage photovoltaic single grid-connected system.

Fig. 3 is the positive-negative sequence double-current ring control block diagram of the method for the invention.

Fig. 4 is the active damping current loop control block diagram under equilibrium condition.

Fig. 5 is the LCL active damping control block diagram under low voltage crossing control strategy.

Fig. 6 is the simplified block diagram of Systematical control

The platform hardware structure chart that Fig. 7 adopts for application the inventive method.

Fig. 8 is the experimental waveform of line voltage under balance is fallen, wherein Fig. 8 a is the voltage current waveform of traditional control method, Fig. 8 b is under 500kW condition, voltage current waveform during grid voltage sags 20%, and Fig. 8 c is the voltage current waveform that three-phase power grid voltage balances moment when dropping to 20%.

Fig. 9 is the experimental waveform of line voltage under imbalance falls condition, wherein Fig. 9 a is the grid-connected voltage current waveform under grid-connected outlet side a phase (single-phase) falls condition, and Fig. 9 b is the grid-connected voltage current waveform under grid-connected outlet side b-c two-phase falls condition.

Embodiment

A kind of low voltage traversing control method of high-power photovoltaic inverter, as shown in Figure 1, the steps include: 1. to consider grid voltage sags especially under imbalance falls condition, the low voltage crossing situation of photovoltaic DC-to-AC converter, calculate forward-order current and negative-sequence current reference set-point, simultaneously for ensureing that photovoltaic DC-to-AC converter has high power factor when grid-connected, choose the reactive power that inverter compensation filter electric capacity absorbs; Wherein, the calculation expression of forward-order current and negative-sequence current reference set-point is:

$\left\{\begin{array}{c}{i}_{\mathrm{inv}\_d}^{p*}=\frac{2e_d^p{P}_0+e_q^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_q}^{p*}=\frac{2e_q^p{P}_0-e_d^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_d}^{n*}=0\\ i_{\mathrm{inv}\_q}^{n*}=0\end{array}\right.$

P in formula ₀, Q ₀be respectively and export meritorious, idle DC component; be respectively the current reference set-point of positive sequence dq axle and negative phase-sequence dq axle; be respectively line voltage positive sequence dq axle component.

The calculation expression choosing the reactive power that inverter compensation filter electric capacity absorbs is:

$\mathrm{\ΔQ}=\mathrm{\ω}{C}_f[{\left(u_{\mathrm{cd}}^p\right)}^2+{\left(u_{\mathrm{cq}}^p\right)}^2+{\left(u_{\mathrm{cq}}^n\right)}^2+{\left(u_{\mathrm{cd}}^n\right)}^2]$

Δ Q is the idle of compensation; ω is the mains frequency detected; C _ffor the filter capacitor in LCL filter; be respectively the positive-negative sequence d axle of filter capacitor, q axle component.

2. based on positive-negative sequence double-current ring control strategy, in modulation signal, filter capacitor current feed-forward is added, by regulating current feed-forward coefficient k _c, the damping realized under uneven condition controls; k _cselection range is 0.2 ~ 0.7, in invention, and preferred k _c=0.5, while effectively can suppressing resonance, also can reduce the impact on the stability of a system.

3. for photovoltaic inverter grid-connected system, transient process during Voltage Drop is considered, design disturbance electric voltage feed forward coefficient k _e, suppress current transient during grid voltage sags to be impacted;

4. the current feed-forward coefficient k that 2. forward-order current 1. calculated according to step and negative-sequence current set-point, step set _cand the disturbance electric voltage feed forward coefficient k that 3. step designs _e, carry out positive-negative sequence double-current ring and control, in order to realize the low voltage crossing of high-power photovoltaic inverter.

Described step 3. in k _eexpression formula be:

$k_e=\frac{1}{G_{\mathrm{inv}}}$

G _invfor the equivalent transfer function of three-phase photovoltaic inverter.

In order to show the beneficial effect of the method for the invention, experiment porch hardware configuration as shown in Figure 7, for 500kW high-power photovoltaic inverter experiment porch hardware structure diagram, mainly comprise electrical network 1, electrical network simulator 7 and high-power photovoltaic inverter major loop 6 and formed.Electrical network simulator 7 is made up of three-phase uncontrollable rectifier bridge 3 and three-phase PWM inverter 4; Photovoltaic DC-to-AC converter major loop is made up of three-phase inverter, LCL filter inductance and DC side filter capacitor; Wherein, connected between electrical network with electrical network simulator by transformer 2, electrical network simulator is connected by isolating transformer 5 with photovoltaic DC-to-AC converter.

The simulating grid programmable power supply that electrical network simulator adopts Taiwan Ai Pusi to produce, model is ACST-L-33800, and its ac output end three-phase voltage, phase angle Independent adjustable, can simulate grid voltage sags during low voltage crossing.Electrical network simulator input gets access to grid through transformer, its DC bus is attached to the direct-flow input end of photovoltaic DC-to-AC converter, the output of photovoltaic DC-to-AC converter is bound up on the ac output end of electrical network simulator by isolating transformer, like this, photovoltaic DC-to-AC converter institute energy requirement is provided by electrical network simulator, and be fed back to electrical network simulator output, realize energy recycling.

For realizing the control of photovoltaic DC-to-AC converter, dissimilar transducer need be added in major loop, wherein, 2 voltage sensors need be added in net side, measuring voltage on line side e _aband e _bc; Filter inductance adds 2 current sensors, measures filter inductance current i _{c_a}, i _{c_b}; Add 2 current sensors at inverter outlet side, measure output current i _{inv_a}, i _{inv_b}.Transducer collection signal is through sampling modulate circuit, send into DSP (the present invention adopt be TMS320LF28335), the control algolithm of DSP is the invention of the design, and the pulse signal that control algolithm calculates is sent to switch tube driving circuit through optical fiber, controls its conducting and shutoff.

Adopt 500kW photovoltaic combining inverter, the parameter of system is as shown in table 1.

Table 1 500kW photovoltaic synchronization inverter system parameter list

Assuming that take traditional restriction amplitude protecting control measure, at grid voltage sags instantaneously by reducing current instruction value to realize LVRT, setting test condition (50kW service conditions is chosen in experiment) is to ensure sufficient safety allowance.Grid voltage sags is set as that three-phase symmetrical drops to 20% rated voltage, and test waveform is as shown in Fig. 8 (a).Under adopting control method of the present invention control grid-connected system to operate in 500kW condition, drop to the test waveform of 20% as shown in Fig. 8 (b), fall under this condition and launch waveform instantaneously as shown in Fig. 8 (c).From Fig. 8 (a), Voltage Drop moment, i _cclearly, and the amplitude of overshoot is much larger than stable operation value for current over pulse phenomenon.If fall under full power condition, overshoot current will cause hardware protection to make net side circuit breaker trip, thus disconnects with electrical network, even has the danger of burning inverter.Compared with Fig. 8 (a), in Fig. 8 (b), stablize the state that is incorporated into the power networks and pass through between state and switch moment, rush of current phenomenon is totally constrained, the meritorious service conditions of rate recovery to fault recovered according to 80kW/s (being greater than 10%) in experiment.Fall the waveform unfolds of moment as shown in Fig. 8 (c), the safe and stable operation of inverter during this procedure ensures that low voltage crossing.

The method of the invention is adopted also to carry out the experiment of line voltage under imbalance falls condition.In experiment porch, electrical network simulator falls fault by regulating three-phase output voltage to simulate imbalance.Test point in experiment selects grid-connected dotted line voltage e _aband e _bcand grid-connected output current i _aand i _b.Test condition is set to single-phasely fall (c phase drops to 20% rated voltage) and two-phase is fallen (a phase drops to 20% with b phase) respectively, and test waveform is respectively as shown in Fig. 9 (a) (single-phase fall), Fig. 9 (b) (two-phase is fallen).From Fig. 9 (a), electrical network simulator c phase voltage drops to 20% rated voltage, through isolating transformer, in voltage on line side, and e _aband e _bcall there is falling in various degree.When two-phase is fallen, it is more obvious that voltage on line side falls the degree of depth, as shown in Fig. 9 (b) voltage waveform.Under different Voltage Drop conditions, the grid-connected current impact falling moment all can be totally constrained.When operating in low voltage crossing state, the negative sequence component of grid-connected current is eliminated, and electric current keeps three-phase equilibrium to export, and has a certain amount of idle output.After line voltage returns to 90% of rated voltage, output current will increase gradually, and active power, with the service conditions of rate recovery to fault of 80kW/s (being greater than 10%), meets the test request of industry.Therefore, the control strategy proposed guarantees that photovoltaic DC-to-AC converter passes through safely subnormal voltage operation interval, exports the functional requirement such as the quality of power supply and reactive power support when control effects meets grid-connected.

The content described that do not elaborate in specification of the present invention belongs to the known prior art of professional and technical personnel in the field.

Claims (1)

1. the low voltage traversing control method of a high-power photovoltaic inverter, it is characterized in that: the steps include: 1. to consider that line voltage is under imbalance falls condition, calculate based on the forward-order current under LCL filter condition and negative-sequence current reference set-point, simultaneously for ensureing that photovoltaic DC-to-AC converter has high power factor when grid-connected, carried out the reactive power that compensation filter electric capacity absorbs by inverter control algorithm;

Founding mathematical models:

$u_{\mathrm{inv}\_x}=L_{\mathrm{inv}}\frac{{\mathrm{di}}_{\mathrm{inv}\_x}}{\mathrm{dt}}+u_{c\_x}$

i _{inv_x}＝i _{g_x}+i _{c_x}

$u_{c\_x}=L_g\frac{{\mathrm{di}}_{g\_x}}{\mathrm{dt}}+e_x$ (1)

In formula: x=a, b, c, wherein a, b, c represent three-phase; u _{inv_x}, i _{inv_x}, e _x, i _{g_x}, u _{c_x}, i _{c_x}represent the voltage and current on a certain phase voltage of inverter outlet side and electric current, line voltage and electric current, filter capacitor respectively; L _invfor inverter side inductance value; L _gfor grid side inductance value;

The power that can obtain inverter output can be expressed as:

$S=\frac{3}{2}(u_{\mathrm{inv}\_\mathrm{dq}}^p{e}^{\mathrm{j\ωt}}+u_{\mathrm{inv}\_\mathrm{dq}}^n{e}^{-\mathrm{j\ωt}})\left(\stackrel{\‾}{i_{\mathrm{inv}\_\mathrm{dq}}^p{e}^{\mathrm{j\ωt}}+i_{\mathrm{inv}\_\mathrm{dq}}^n{e}^{-\mathrm{j\ωt}}}\right)---\left(2\right)$

Active power of output and reactive power can be expressed as:

P＝(P ₀+P _c2cos2ωt+P _s2sin2ωt) (3)

Q＝(Q ₀+Q _c2cos2ωt+Q _s2sin2ωt) (4)

In formula: P ₀, Q ₀be respectively and export meritorious, idle DC component; be respectively inverter outlet side voltage positive-negative sequence d axle, q axle component; be respectively inverter outlet side electric current positive-negative sequence d axle, q axle component; P _c2, P _s2, Q _c2, Q _s2for exporting meritorious, idle wave component;

ω is line voltage angular frequency, real component and idle component is expanded into:

$\left[\begin{array}{c}{P}_0\\ P_{c2}\\ P_{s2}\\ Q_0\\ Q_{c2}\\ Q_{s2}\end{array}\right]=\frac{3}{2}\left[\begin{array}{cccc}{u}_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^p& u_{\mathrm{inv}\_d}^n& u_{\mathrm{inv}\_q}^n\\ u_{\mathrm{inv}\_d}^n& u_{\mathrm{inv}\_q}^n& u_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^p\\ u_{\mathrm{inv}\_q}^n& -u_{\mathrm{inv}\_d}^n& -u_{\mathrm{inv}\_q}^p& u_{\mathrm{inv}\_d}^p\\ u_{\mathrm{inv}\_q}^p& -u_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^n& -u_{\mathrm{inv}\_d}^n\\ u_{\mathrm{inv}\_q}^n& -u_{\mathrm{inv}\_d}^n& u_{\mathrm{inv}\_q}^p& -u_{\mathrm{inv}\_d}^p\\ -u_{\mathrm{inv}\_d}^n& -u_{\mathrm{inv}\_q}^n& u_{\mathrm{inv}\_d}^p& u_{\mathrm{inv}\_q}^p\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{inv}\_d}^p\\ i_{\mathrm{inv}\_q}^p\\ i_{\mathrm{inv}\_d}^n\\ i_{\mathrm{inv}\_q}^n\end{array}\right]---\left(5\right)$

In order to utilize this known measuring amount of line voltage, obtaining accurate instruction current, need convert formula (5), carrying out coordinate transform by formula (1), and carry out positive-negative sequence and be separated the Mathematical Modeling that can obtain under positive-negative sequence component and can be expressed as:

$\left\{\begin{array}{c}{L}_{\mathrm{inv}}\frac{{\mathrm{di}}_{\mathrm{inv}\_\mathrm{dq}}^p}{\mathrm{dt}}=u_{\mathrm{inv}\_\mathrm{dq}}^p-u_{c\_\mathrm{dq}}^p+\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_\mathrm{qd}}^p\\ L_g\frac{{\mathrm{di}}_{g\_\mathrm{dq}}^p}{\mathrm{dt}}=u_{c\_\mathrm{dq}}^p-e_{\mathrm{dq}}^p\±\mathrm{\ω}{L}_g{i}_{g\_\mathrm{qd}}^p\\ C_f\frac{{\mathrm{du}}_{c\_\mathrm{dq}}^p}{\mathrm{dt}}=i_{\mathrm{inv}\_\mathrm{dq}}^p-i_{g\_\mathrm{dq}}^p\±\mathrm{\ω}{C}_f{u}_{c\_\mathrm{qd}}^p\end{array}\right.---\left(6\right)$

In formula, C _ffor the filter capacitor in LCL filter;

Can be obtained by formula (6), under systematic steady state condition, the pass between inverter side voltage and line voltage is:

$\begin{array}{c}{u}_{\mathrm{inv}\_d}^p=e_d^p-\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p\\ u_{\mathrm{inv}\_q}^p=e_q^p+\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p\\ u_{\mathrm{inv}\_d}^n=e_d^n+\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n\\ u_{\mathrm{inv}\_q}^n=e_q^n-\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n\end{array}---\left(7\right)$

Bring formula (7) into formula (5) can obtain

$\begin{array}{c}\left[\begin{array}{c}{P}_0\\ Q_0\\ P_{c2}\\ P_{s2}\end{array}\right]=\frac{3}{2}\left[\begin{array}{cccc}{e}_d^p& e_q^p& e_d^n& e_q^n\\ e_q^p& -e_d^p& e_q^n& -e_d^n\\ e_d^n& e_q^n& e_d^p& e_q^p\\ e_q^n& -e_d^n& -e_q^p& e_d^p\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{inv}\_d}^p\\ i_{\mathrm{inv}\_q}^p\\ i_{\mathrm{inv}\_d}^n\\ i_{\mathrm{inv}\_q}^n\end{array}\right]+\\ \frac{3}{2}\left[\begin{array}{cccc}-{\mathrm{\ω}}_{\mathrm{inv}}{\mathrm{Li}}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p& \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p& \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n\\ \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p& \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n\\ \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n+\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p& \mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p+\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p\\ -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^n-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^n& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_d}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gd}}^p& -\mathrm{\ω}{L}_{\mathrm{inv}}{i}_{\mathrm{inv}\_q}^p-\mathrm{\ω}{L}_g{i}_{\mathrm{gq}}^p\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{inv}\_d}^p\\ i_{\mathrm{inv}\_q}^p\\ i_{\mathrm{inv}\_d}^n\\ i_{\mathrm{inv}\_q}^n\end{array}\right]\end{array}---\left(8\right)$

Formula (8) is the relation between known measuring amount and inverter output power, for eliminating negative-sequence current namely with in formula (8) for target, calculating, ignoring this impact for simplifying, therefore the calculation expression of forward-order current and negative-sequence current reference set-point is:

$\left\{\begin{array}{c}{i}_{\mathrm{inv}\_d}^{p^{*}}=\frac{2e_d^p{P}_0+e_q^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_q}^{p^{*}}=\frac{2e_q^p{P}_0-e_d^p{Q}_0}{3[{\left(e_d^p\right)}^2+{\left(e_q^p\right)}^2]}\\ i_{\mathrm{inv}\_d}^{n^{*}}=0\\ i_{\mathrm{inv}\_q}^{n^{*}}=0\end{array}\right.$

P in formula ₀, Q ₀be respectively and export meritorious, idle DC component; be respectively the current reference set-point of positive sequence d axle, q axle and negative phase-sequence d axle, q axle; be respectively line voltage positive sequence d axle and q axle component;

The calculation expression choosing the reactive power that inverter control algorithm compensation filter electric capacity absorbs is:

$\mathrm{\ΔQ}=\mathrm{\ω}{C}_f[{\left(u_{\mathrm{cd}}^p\right)}^2+{\left(u_{\mathrm{cq}}^p\right)}^2+{\left(u_{\mathrm{cq}}^n\right)}^2+{\left(u_{\mathrm{cd}}^n\right)}^2]$

Δ Q is the idle of compensation; ω is the mains frequency detected; C _ffor the filter capacitor in LCL filter; be respectively the positive-negative sequence d of filter capacitor, q axle component;

2. based on positive-negative sequence double-current ring control strategy, design and add filter capacitor current feed-forward signal in modulation signal, by regulating current feed-forward coefficient k _c, control with the active damping realized under unbalance control condition; Current feed-forward coefficient k _cselection range be 0.2 ~ 0.7, while effectively can suppressing resonance, also can reduce the impact on the stability of a system;

3. for transient current surge during Voltage Drop, design disturbance electric voltage feed forward coefficient k _e, transient suppression impacts, disturbance electric voltage feed forward coefficient k _eaccount form

$k_e=\frac{1}{G_{\mathrm{inv}}}$

G _invfor the equivalent transfer function of three-phase photovoltaic inverter;

4. the current feed-forward coefficient k that 2. forward-order current 1. calculated according to step and negative-sequence current set-point, step design _cand the disturbance electric voltage feed forward coefficient k that 3. step designs _e, carry out positive-negative sequence double-current ring and control, in order to realize the low voltage crossing of high-power photovoltaic inverter.