CN103346689B - Grid-connected inverter parallel system circulating current restraining method under imbalance condition of inductance - Google Patents

Abstract

The invention relates to a grid-connected inverter parallel system circulating current restraining method under an imbalance condition of inductance, relates to the technical field of the circulating current restraining of grid-connected inverter parallel systems, and aims at solving the problems in the grid-connected inverter parallel system that when the three-phase inductance of the parallel inverter is varied or imbalanced, the circulating current is still large through using the existing circulating current restraining method. Since the circulating current flowing between every two inverters is identical in value and opposite in directions, the circulating current between every two inverters can be restrained only by restraining the circulating current of one inverter, so that a purpose for restraining the system circulating current can be achieved. Under the imbalance of the three-phase inductance of the inverter, compared with a PI control method, for the method, the peak value of the circulating current can be reduced by 40 to 65 percent. The grid-connected inverter parallel system circulating current restraining method is suitable for restraining the circulating current in the grid-connected inverter parallel system.

Description

Combining inverter parallel system circulation inhibition method under inductance unbalance condition

Technical field

The invention belongs to the loop current suppression technical field of combining inverter parallel system.

Background technology

Three-phase PWM inverter, because its AC output current is sinusoidal, power factor is high, current distortion is little, receives much concern in distributed energy electricity generation system.Along with the increase of distributed energy electricity generation system capacity, the power grade of inverter is had higher requirement.In order to ensure the reliability of system at the power requirement meeting system simultaneously, reduce cost and the volume of system simultaneously, often direct Parallel Control is carried out to inverter, be directly together in parallel by the DC side of inverter and AC.This parallel-connection structure increase power system capacity while, also for circulation provides path.Circulation can have a negative impact to system, as made waveform distort, increasing system loss, reducing system effectiveness, even exceeding the power grade etc. of equipment, needs to consider to suppress circulation when therefore controlling parallel system.Circulation inhibition method conventional at present has:

Isolation is in parallel.This method can use isolating transformer or DC side to adopt independent direct current source to realize by AC, simple, but can increase cost and the volume of system.

Circulation flow path uses interphase reactor, suppresses circulation in order to form high impedance.This method has good inhibition for high frequency circulating currents, and for the medium and low frequency composition in circulation, inhibition is limited.

Staggered intermittent space vector modulating method.This method effectively can reduce the THD of system, but can increase the switching loss of system, reduces system effectiveness, especially in large-power occasions.

PI controls circulation.When the given electric current of each shunt chopper is equal, this method can obtain good control effects, and implements fairly simple.But PI controller can only suppress the circulation existed in this cycle, for the circulation being about in next cycle produce, its control action is very limited.When the given electric current of each inverter does not wait or ac filter inductance parameters is different, dynamic response is comparatively slow, adopts this control method can make circulation peak-to-peak value decline 54% (relevant with carrying out practically condition), controls circulation weak effect.

PI voltage zero vector feedforward controls circulation.Greatly can improve PI control algolithm in each shunt chopper deficiency when given electric current does not wait or inductance parameters is different by introducing zero vector feedforward, and there is good dynamic response, volume and the cost of system can not be increased.

Summary of the invention

The present invention is to solve in combining inverter parallel system, when the change of shunt chopper three pole reactor or imbalance, and the problem that when method of existing suppression circulation controls, circulation is still larger.Combining inverter parallel system circulation inhibition method under a kind of inductance unbalance condition is now provided.

Combining inverter parallel system circulation inhibition method under inductance unbalance condition, combining inverter in described combining inverter parallel system is common DC bus, the direct structure in parallel of AC, described parallel system adopts PI method to control, for the circulation inhibition method of above-mentioned combining inverter parallel system, the method is in two inverter parallel systems, the circulation of an inverter of combining inverter is controlled, specifically comprises the following steps:

Step one: to the zero-sequence current i of the second inverter (2) _z2sample, then perform step 2;

Step 2: utilize zero-sequence current PI controller to the zero-sequence current i of the second inverter (2) _z2with the zero-sequence current set-point i of the second inverter (2) _{z2_ref}do difference, and using the input signal of this difference as zero-sequence current PI controller, utilize the PI algorithm of zero-sequence current PI controller to regulate this input signal, obtain PI regulated value, then perform step 3 and step 4 simultaneously;

Step 3: by the difference DELTA d of the non-zero duty ratio of the space vector pulse width modulation of the first inverter (1) and the second inverter (2) ₁₂divided by 12, obtain the regulated quantity of non-zero, then perform step 5;

Step 4: by the difference u of the inductance residual voltage of the first inverter (1) and the second inverter (2) _lz12divided by the DC bus-bar voltage u of 2 times _dc, obtain the regulated quantity of inductance residual voltage, then perform step 5;

Step 5: the regulated quantity of the inductance residual voltage that the regulated quantity of the non-zero that the PI regulated value obtain step 2 and step 3 obtain and step 4 obtain is poor, the difference obtained is exported by zero sequence circulation controller 3, and this difference is as the correction value y of the second inverter (2) zero vector ₂, then perform step 6;

Step 6: the correction value y of the second inverter (2) zero vector utilizing step 5 to obtain ₂the distribution of zero vector in the space vector pulse width modulation of the second inverter (2) is regulated in real time, completes the suppression to circulation.

The zero-sequence current i of the second inverter (2) in above-mentioned steps one _z2for:

$i_{z2}=\frac{i_{a2}+i_{b2}+i_{c2}}{3}---\left(1\right)$

I in formula _k2(k=a, b, c) is respectively a phase, b phase, the c phase current of the second inverter (2).

Δ d described in above-mentioned steps three ₁₂for:

Δd ₁₂=(d ₂₁-d ₁₁)-(d ₂₂-d ₁₂) （2）

D in formula ₁₁be the duty ratio of first non-zero of the first inverter, d ₂₁be the duty ratio of second non-zero of the first inverter, d ₁₂be the duty ratio of first non-zero of the second inverter (2), d ₂₂it is the duty ratio of second non-zero of the second inverter (2).U described in above-mentioned steps four _lz12for:

$u_{L_{z12}}=u_{L_{z1}}-u_{L_{z2}}$

$=\frac{\mathrm{\ω}{i}_{d2}(L_{a2}\sin \mathrm{\ωt}+L_{b2}\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c2}\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}$

$-\frac{\mathrm{\ω}{i}_{d1}(L_{a1}\sin \mathrm{\ωt}+L_{b1}\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c1}\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}---\left(3\right)$

$+\frac{\mathrm{\ω}{i}_{q2}(L_{a2}\cos \mathrm{\ωt}+L_{b2}\cos (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c2}\cos (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}$

$-\frac{\mathrm{\ω}{i}_{q1}(L_{a1}\cos \mathrm{\ωt}+L_{b1}\cos (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c1}\cos (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}$

In formula be the inductance residual voltage of the first inverter (1), be the inductance residual voltage of the second inverter (2), i _dx(x=1,2) are respectively the d axle component of the three-phase current of the first inverter (1) and the second inverter (2), i _qx(x=1,2) are respectively the q axle component of the three-phase current of the first inverter (1) and the second inverter (2), L _kx(k=a, b, c; X=1,2) be respectively the inductance value of the first inverter (1) and second inverter a, b, c phase, ω is the angular frequency of electrical network.

The correction value y of the second inverter (2) zero vector in above-mentioned steps five ₂for:

$y_2=K_{p\_z}\·(i_{z2\_\mathrm{ref}}-i_{z2})+K_{i\_z}\∫(i_{z2\_\mathrm{ref}}-i_{z2})\mathrm{dt}-\frac{\mathrm{\Δ}{d}_{12}}{12}-\frac{u_{\mathrm{Lz}12}}{2u_{\mathrm{dc}}}---\left(4\right)$

K in formula _{p_z}for the proportionality coefficient of zero-sequence current PI controller, K _{i_z}for the integral coefficient of zero-sequence current PI controller.

The concrete grammar of the real-time adjustment described in above-mentioned steps six is:

In a control cycle T by the control of the on off state of the A phase of the second inverter (2) be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

In same control cycle T by the control of the on off state of the B phase of the second inverter (2) be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

In same control cycle T by the control of the on off state of the C phase of the second inverter (2) be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

D ₁₁be the duty ratio of first non-zero of the first inverter, d ₂₁be the duty ratio of second non-zero of the first inverter, d ₁₂be the duty ratio of first non-zero of the second inverter (2), d ₂₂be the duty ratio of second non-zero of the second inverter (2), d ₀₂it is the duty ratio of the second inverter (2) zero vector.

Combining inverter parallel system circulation inhibition method under inductance unbalance condition of the present invention, owing to flowing through circulation equal and opposite in direction between any two inverters, direction is contrary, only need to suppress the circulation of one of them inverter, circulation between can realizing any two inverters suppresses, and then reaches the object of suppression system circulation.When inverter three pole reactor is uneven, compare PI control method, circulation peak-to-peak value can reduce 40% to 65%.

Accompanying drawing explanation

Fig. 1 is the flow chart of combining inverter parallel system circulation inhibition method under inductance unbalance condition.

When Fig. 2 is for two inverter parallel structures, the topological structure schematic diagram of system; Wherein l _za, l _zb, l _zca circulation flow path of a, b, c phase in the corresponding parallel-connection structure of difference, every bar circulation flow path comprises upper brachium pontis and the inverter 2k (k=a of k (k=a, b, c) phase in inverter 1 respectively, b, c) the lower brachium pontis of phase, a, b, c phase respectively has a circulation flow path not draw in addition, and these three circulation bar paths comprise k (k=a in inverter 1 respectively, b, c) the lower brachium pontis of phase and the upper brachium pontis of inverter 2k (k=a, b, c) phase.

Fig. 3 is the principle schematic of combining inverter parallel system circulation inhibition method under inductance unbalance condition.

Fig. 4 is the equivalent-circuit model schematic diagram of the zero sequence circulation flow path of combining inverter.

Fig. 5 is the schematic diagram of the space vector pulse width modulation of inverter.

Fig. 6 is that zero vector correction value y introduced by the second inverter 2 ₂after space vector distribution map.

Fig. 7 is the control block diagram of zero-sequence current ring when adopting PI to control.

Fig. 8 adopts the control block diagram based on the zero-sequence current ring of voltage zero vector feedforward.

Fig. 9 is the zero sequence circulation control block diagram of combining inverter parallel system circulation inhibition method under inductance unbalance condition.

Embodiment

Embodiment one: illustrate present embodiment with reference to Fig. 1 and Fig. 3, combining inverter parallel system circulation inhibition method under inductance unbalance condition described in present embodiment, combining inverter in described combining inverter parallel system is common DC bus, the direct structure in parallel of AC, described parallel system adopts PI method to control, for the circulation inhibition method of above-mentioned combining inverter parallel system, the method is in two inverter parallel systems, the circulation of an inverter of combining inverter is controlled, specifically comprises the following steps:

Step one: to the zero-sequence current i of the second inverter 2 _z2sample, then perform step 2;

Step 2: utilize zero-sequence current PI controller to the zero-sequence current i of the second inverter 2 _z2with the zero-sequence current set-point i of the second inverter 2 _{z2_ref}do difference, and using the input signal of this difference as zero-sequence current PI controller, utilize the PI algorithm of zero-sequence current PI controller to regulate this input signal, obtain PI regulated value, then perform step 3 and step 4 simultaneously;

Step 3: by the difference DELTA d of the non-zero duty ratio of the space vector pulse width modulation of the first inverter 1 and the second inverter 2 ₁₂divided by 12, obtain the regulated quantity of non-zero, then perform step 5;

Step 4: by the difference u of the inductance residual voltage of the first inverter 1 and the second inverter 2 _lz12divided by the DC bus-bar voltage u of 2 times _dc, obtain the regulated quantity of inductance residual voltage, then perform step 5;

Step 5: the regulated quantity of the inductance residual voltage that the regulated quantity of the non-zero that the PI regulated value obtain step 2 and step 3 obtain and step 4 obtain is poor, the difference obtained is exported by zero sequence circulation controller 3, and this difference is as the correction value y of the second inverter 2 zero vector ₂, then perform step 6;

Step 6: the correction value y of the second inverter 2 zero vector utilizing step 5 to obtain ₂the distribution of zero vector in the space vector pulse width modulation of the second inverter 2 is regulated in real time, completes the suppression to circulation.

Combining inverter parallel system circulation inhibition method under inductance unbalance condition of the present invention, owing to flowing through circulation equal and opposite in direction between any two inverters, direction is contrary, only need to suppress the circulation of one of them inverter, circulation between can realizing any two inverters suppresses, and then reaches the object of suppression system circulation.

Embodiment two: present embodiment is described further combining inverter parallel system circulation inhibition method under the inductance unbalance condition described in embodiment one, in present embodiment, the zero-sequence current i of the second inverter 2 in step one _z2for:

$i_{z2}=\frac{i_{a2}+i_{b2}+i_{c2}}{3}---\left(1\right)$

I in formula _k2(k=a, b, c) is respectively a phase, b phase, the c phase current of the second inverter 2.

Embodiment three: present embodiment is described further combining inverter parallel system circulation inhibition method under the inductance unbalance condition described in embodiment one, in present embodiment, the d of Δ described in step 3 ₁₂for:

Δd ₁₂=(d ₂₁-d ₁₁)-(d ₂₂-d ₁₂) （2）

D in formula ₁₁be the duty ratio of first non-zero of the first inverter, d ₂₁be the duty ratio of second non-zero of the first inverter, d ₁₂be the duty ratio of first non-zero of the second inverter 2, d ₂₂it is the duty ratio of second non-zero of the second inverter 2.

Embodiment four: present embodiment is described further combining inverter parallel system circulation inhibition method under the inductance unbalance condition described in embodiment one, in present embodiment, u described in step 4 _lz12for:

$u_{L_{z12}}=u_{L_{z1}}-u_{L_{z2}}$

$=\frac{\mathrm{\ω}{i}_{d2}(L_{a2}\sin \mathrm{\ωt}+L_{b2}\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c2}\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}$

$-\frac{\mathrm{\ω}{i}_{d1}(L_{a1}\sin \mathrm{\ωt}+L_{b1}\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c1}\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}---\left(3\right)$

$+\frac{\mathrm{\ω}{i}_{q2}(L_{a2}\cos \mathrm{\ωt}+L_{b2}\cos (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c2}\cos (\mathrm{\ω}+\frac{2}{3}\mathrm{\π}))}{3}$

$-\frac{\mathrm{\ω}{i}_{q1}(L_{a1}\cos \mathrm{\ωt}+L_{b1}\cos (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c1}\cos (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}$

In formula be the inductance residual voltage of the first inverter 1, be the inductance residual voltage of the second inverter 2, i _dx(x=1,2) are respectively the d axle component of the three-phase current of the first inverter 1 and the second inverter 2, i _qx(x=1,2) are respectively the q axle component of the three-phase current of the first inverter 1 and the second inverter 2, L _kx(k=a, b, c; X=1,2) be respectively the inductance value of the first inverter 1 and second inverter a, b, c phase, ω is the angular frequency of electrical network.Embodiment five: present embodiment is described further combining inverter parallel system circulation inhibition method under the inductance unbalance condition described in embodiment one, in present embodiment, the correction value y of the second inverter 2 zero vector in step 5 ₂for:

$y_2=K_{p\_z}\·(i_{z2\_\mathrm{ref}}-i_{z2})+K_{i\_z}\∫(i_{z2\_\mathrm{ref}}-i_{z2})\text{dt-}\frac{\mathrm{\Δ}{d}_{12}}{12}-\frac{u_{\mathrm{Lz}12}}{2u_{\mathrm{dc}}}---\left(4\right)$

K in formula _{p_z}for the proportionality coefficient of zero-sequence current PI controller, K _{i_z}for the integral coefficient of zero-sequence current PI controller.

Embodiment six: present embodiment is described further combining inverter parallel system circulation inhibition method under the inductance unbalance condition described in embodiment one, and in present embodiment, the concrete grammar of the real-time adjustment described in step 6 is:

In a control cycle T by the control of the on off state of the A phase of the second inverter 2 be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

In same control cycle T by the control of the on off state of the B phase of the second inverter 2 be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

In same control cycle T by the control of the on off state of the C phase of the second inverter 2 be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

D ₁₁be the duty ratio of first non-zero of the first inverter, d ₂₁be the duty ratio of second non-zero of the first inverter, d ₁₂be the duty ratio of first non-zero of the second inverter 2, d ₂₂be the duty ratio of second non-zero of the second inverter 2, d ₀₂it is the duty ratio of the second inverter 2 zero vector.In the present invention, three-phase grid-connected inverter in parallel adopts common DC bus and the direct structure in parallel of AC, for two inverter parallel structures, as shown in Figure 2, parallel system DC bus capacitor is 2C, C is single three-phase grid-connected inverter DC bus capacitor, and two inverter module power are equal.This topological structure is that circulation provides path, and circulation flow path has 6, uses l in Fig. 2 _za, l _zb, l _zcillustrating wherein 3 circulation flow paths, needing when carrying out Controller gain variations to suppress circulation.

For each shunt chopper, a, b, c three-phase respectively has two circulation flow paths, and circulation is uniformly distributed in each phase, and the circulation flowing through inverter is 3 times of each phase circulation.The circulation suppressing to suppress inverter is carried out by the mean value in each phase of the circulation to inverter.

When considering each inverter three pole reactor to the affecting of circulation, for two inverter parallel structures, the Mathematical Modeling of parallel system under two-phase synchronous rotating frame can be expressed as:

$\left\{\begin{array}{c}(L_{b1}+\frac{L_{\cos 2\mathrm{nl}}}{3}-\frac{L_{\sin 2p1}^2}{9L_{b1}+3L_{\sin 2n1}})\frac{{\mathrm{di}}_{d1}}{\mathrm{dt}}-\mathrm{\ω}(L_{b1}+\frac{L_{\cos 2n1}}{3}-\frac{L_{\sin 2p1}^2}{9L_{b1}+3L_{\sin 2n1}})i_{q1}\\ =u_{d1}-e_d+\frac{L_{\sin 2p1}}{3L_{b1}+L_{\sin 2n1}}(u_{q1}-e_q)\\ (L_{b1}+\frac{L_{\sin 2n1}}{3}-\frac{L_{\sin 2p1}^2}{9L_{b1}+3L_{\cos 2n1}})\frac{{\mathrm{di}}_{q1}}{\mathrm{dt}}-\mathrm{\ω}(L_{b1}+\frac{L_{\sin 2n1}}{3}-\frac{L_{\sin 2p1}^2}{9L_{b1}+3L_{\cos 2n1}})i_{d1}\\ =u_{q1}-e_q+\frac{L_{\sin 2p1}}{3L_{b1}+L_{\cos 2n1}}(u_{d1}-e_d)\end{array}\right.---\left(5\right)$

$\left\{\begin{array}{c}(L_{b2}+\frac{L_{\cos 2n2}}{3}-\frac{L_{\sin 2p2}^2}{{9L}_{b2}+3L_{2\sin 2n}})\frac{d{i}_{d2}}{\mathrm{dt}}-\mathrm{\ω}(L_{b2}+\frac{L_{\cos 2n2}}{3}-\frac{L_{\sin 2p2}^2}{9L_{b2}+3L_{\sin 2n}})i_{q1}\\ =u_{d2}-e_d+\frac{L_{\sin 2p2}}{3L_{b2}+L_{\sin 2n2}}(u_{q2}-e_q)\\ (L_{b2}+\frac{L_{\sin 2n2}}{3}-\frac{L_{\sin 2p2}^2}{9L_{b2}+3L_{\cos 2n2}})\frac{d{i}_{q2}}{\mathrm{dt}}-\mathrm{\ω}(L_{b2}+\frac{L_{\sin 2n2}}{3}-\frac{L_{\sin 2p2}^2}{9L_{b2}+3L_{\cos 2n2}})i_{d2}\\ =u_{q2}-e_q+\frac{L_{\sin 2p2}}{3L_{b2}+L_{\cos 2n2}}(u_{d2}-e_d)\end{array}\right.----\left(6\right)$

$\left\{\begin{array}{c}{u}_{\mathrm{dx}}=d_{\mathrm{dx}}{u}_{\mathrm{dc}}\\ u_{\mathrm{qx}}=d_{\mathrm{qx}}{u}_{\mathrm{dc}}\end{array}\right.(x=\mathrm{1,2})---\left(7\right)$

$(L_{m1}+L_{m2})\frac{d_{\mathrm{iz}2}}{\mathrm{dt}}=u_{\mathrm{Lz}12}+\mathrm{\Δ}{d}_z{u}_{\mathrm{dc}}---\left(8\right)$

Wherein:

$L_{\cos 2\mathrm{nx}}=\frac{1}{2}{L}_{\mathrm{ax}}\cos 2\mathrm{\ωt}+\frac{1}{2}{L}_{\mathrm{bx}}\cos (2\mathrm{\ωt}+\frac{2}{3}\mathrm{\π})+\frac{1}{2}{L}_{\mathrm{cx}}\cos (2\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+\frac{L_{\mathrm{ax}}+L_{\mathrm{cx}}+2L_{\mathrm{bx}}}{2}$

$L_{\sin 2\mathrm{px}}=L_{\mathrm{ax}}\sin 2\mathrm{\ωt}+L_{\mathrm{cx}}\sin (2\mathrm{\ωt}-\frac{2\mathrm{\π}}{3})+L_{\mathrm{bx}}\sin (2\mathrm{\ωt}+\frac{2\mathrm{\π}}{3})$

$L_{\sin 2\mathrm{nx}}=-\frac{1}{2}{L}_{\mathrm{ax}}\cos 2\mathrm{\ωt}-\frac{1}{2}{L}_{\mathrm{bx}}\cos (2\mathrm{\ωt}+\frac{2}{3}\mathrm{\π})-\frac{1}{2}{L}_{\mathrm{cx}}\cos (2\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+\frac{L_{\mathrm{ax}}+L_{\mathrm{cx}}-2L_{\mathrm{bx}}}{2}$

$L_{\mathrm{mx}}=\frac{L_{\mathrm{ax}}+L_{\mathrm{bx}}+L_{\mathrm{cx}}}{3}$

$u_{\mathrm{Lzx}}=\frac{\mathrm{\ω}{i}_{\mathrm{dx}}(L_{\mathrm{ax}}\sin \mathrm{\ωt}+L_{\mathrm{bx}}\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{\mathrm{cx}}\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}$

$-\frac{\mathrm{\ω}{i}_{\mathrm{qx}}(L_{\mathrm{ax}}\cos \mathrm{\ωt}+L_{\mathrm{bx}}\cos (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{\mathrm{cx}}\cos (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}$

X=1, corresponding first inverter 1 and the second inverter 2, i of 2 difference _dxfor the d axle component of each inverter three-phase current, i _qxfor the q axle component of each inverter three-phase current, L _kx(k=a, b, c; X=1,2) be the inductance value of each phase of each inverter, e _dfor the d axle component of line voltage, e _qfor the q axle component of line voltage, ω is the angular frequency of electrical network, d _dxfor the d axle component of each inverter three-phase duty ratio, d _qxfor the q axle component of each inverter three-phase duty ratio, u _dcfor DC bus-bar voltage, i _zxfor the zero-sequence current of each inverter, Δ d _zbe the difference of two inverter zero sequence duty ratios, Δ d _z=d _z1-d _z2, d _zxbe respectively the zero sequence duty ratio of two inverters.

Like this, the zero sequence circulation flow path equivalent-circuit model schematic diagram of combining inverter is obtained, as shown in Figure 4.

According to the Mathematical Modeling of the zero-sequence current of three-phase grid-connected inverter in parallel, i.e. formula (8), can infer that the rate of change of the zero-sequence current between two inverters is determined jointly by the difference of the difference of the zero sequence duty ratio of two inverters and the residual voltage of ac filter inductance.Combining inverter parallel system circulation inhibition method under inductance unbalance condition, owing to flowing through circulation equal and opposite in direction between any two inverters, direction is contrary, only need to suppress the circulation of one of them inverter, circulation between can realizing any two inverters suppresses, and then reaches the object of suppression system circulation.

For inverter parallel system, because zero axle is a undamped loop only containing inductance, in systems in practice, three kinds of situations more often occur below:

The given electric current of two inverters is unequal;

The inductance difference (each inverter three pole reactor balance) of two inverters;

The three pole reactor change of each inverter or imbalance cause its inductance residual voltage there are differences.

Any one condition all may cause the zero sequence duty ratio of two inverters there are differences, even if its difference is less, also can make to form larger zero-sequence current between inverter.Therefore, during inverter parallel, need to consider the suppression to zero-sequence component.Can find out simultaneously, in the Circulation Model of inverter parallel structure, the difference of the inductance residual voltage of any two inverters can to the interference of zero-axis current ring, and interference is generally not easily got rid of, but the circulation between inverter is directly related with the difference of the zero sequence duty ratio of two inverters, and zero sequence duty ratio is a controlled variable, the suppression of circulation between therefore can being realized inverter by the difference of the zero sequence duty ratio regulating two inverters in each control cycle.

In three-phase grid-connected inverter, usually adopt SVPWM mode, this modulation system adopts two non-zero V usually _i(i=1,2,3,4,5,6) and zero vector V _i(i=0,7) synthesize control vector, vector V _ithe definition of (i=0 ~ 7), as shown in Figure 5.For two inverter parallel structures, if the duty ratio of the second inverter 2 two non-zero is respectively d ₁₂, d ₂₂, zero vector duty ratio is d ₀₂, then:

d ₀₂=1-d ₁₂-d ₂₂(9)

Different modulator approaches can obtain the allocation result of zero vector, thus can change duty ratio and the zero sequence duty ratio of each each phase of shunt chopper, but the difference of the duty ratio of two-phase can not change arbitrarily.Therefore the control objectives of system, namely ac-side current and DC bus-bar voltage can not be affected.Just zero sequence duty ratio d can be controlled by the distribution controlling zero vector _z, thus control zero-sequence current.For the exemplary sector shown in Fig. 5, within a PWM cycle, if zero vector V ₇action time be zero vector V ₀action time be as shown in Figure 6, zero vector correction value y introduced by the second inverter 2 ₂after space vector distribution map in variable y ₂meet:

$-\frac{d_{02}}{4}\≤y_2\≤\frac{d_{02}}{4}---\left(10\right)$

Therefore zero vector V ₀, V ₇action time regulate by its duty ratio, the span of both duty ratios is [0, d ₀₂], and both sums are d ₀₂.Now,

$d_{z2}=\frac{(d_{a2}+d_{b2}+d_{c2})}{3}$

$=\frac{(d_{12}+d_{22}+\frac{d_{02}}{2}-2y_2)+(d_{22}+\frac{d_{02}}{2}-2y_2)+(\frac{d_{02}}{2}-2y_2)}{3}---\left(11\right)$

$=\frac{(d_{12}+2d_{22}+\frac{3}{2}{d}_{02}-6y_2)}{3}$

To the first inverter 1, formula (11) is set up equally.Therefore the difference of the zero sequence duty ratio of two inverters is:

$\mathrm{\Δ}{d}_z=d_{z1}-d_{z2}=\frac{(d_{11}+2d_{21}+\frac{3}{2}{d}_{01}-6y_1)}{3}-\frac{(d_{12}+2d_{22}+\frac{3}{2}{d}_{02}-6y_2)}{3}---\left(12\right)$

Due to the circulation between any two inverters, equal and opposite in direction in two inverters, direction is contrary, the carrying out of one of them inverter is controlled, and then can the circulation of control system, therefore, for two inverter parallel structures, make the zero vector correction value y of the first inverter 1 ₁=0.In addition, due to d _0x=1-d _1x-d _2x(x=1,2), formula (12) can abbreviation be:

$\mathrm{\Δ}{d}_z=\frac{1}{6}(-d_{11}+d_{21}+d_{12}-d_{22}+12y_2)---\left(13\right)$

I.e. △ d ₁₂=-d ₁₁+ d ₂₁+ d ₁₂-d ₂₂, then above formula can turn to:

$\mathrm{\Δ}{d}_z=\frac{1}{6}(\mathrm{\Δ}{d}_{12}+12y_2)---\left(14\right)$

For two inverter parallel systems, ignore the difference of each inverter inductor residual voltage, when the given electric current of two inverters is equal, the voltage given value that current regulator exports is substantially equal, therefore d ₁₁=d ₁₂, d ₂₁=d ₂₂, now, Δ d ₁₂=0, therefore

Δd _z=2y ₂(15)

Therefore, the Mathematical Modeling of zero-sequence current under synchronous rotating frame, namely formula (8) can turn to:

$(L_{m1}+L_{m2})\frac{d{i}_{z2}}{\mathrm{dt}}=2y_2\·u_{\mathrm{dc}}---\left(16\right)$

Assuming that u _dckeep constant, Laplace transform done to above formula, can obtain:

s(L _m1+L _m2)I _z2=2U _dcY ₂(17)

Y in above formula ₂, I _z2be respectively variable y ₂, i _z2laplace transformation.

As can be seen from the above equation, zero axle of each inverter and d axle and q axle full decoupled, and zero axle is a first-order system, therefore, it is very high that the bandwidth of zero-sequence current ring can design, and pi regulator can be adopted as the controller of zero-sequence current, by the set-point of zero-sequence current and sampled value poor, PI adjustment is carried out to its deviation, the correction value of the second inverter 2 zero vector can be obtained:

y ₂=K _{p_z}·(i _{z2_ref}-i _z2)+K _{i_z}∫(i _{z2_ref}-i _z2)dt (18)

The control block diagram of corresponding zero-sequence current ring as shown in Figure 7.

To there is circulation, PI controller can play regulatory role, but this control cycle is about to the circulation produced, PI controls to suppress in time, dynamic response is slower, therefore only when the given electric current of the inverter of each parallel connection is equal and two inverter inductor residual voltages are identical, good control effects can be obtained, when the given electric current of two inverters is unequal the inductance difference of each inverter (each inverter three pole reactor balance) or each inverter three pole reactor imbalance causes the inductance zero sequence of two inverters unequal time, its circulation control effects is poor.

Ignore the difference of two inverter inductor residual voltages, consider that the given electric current of two inverters is unequal, the unequal situation of filter inductance, can obtain according to formula (13):

$(L_{m1}+L_{m2})\frac{d{i}_{z2}}{\mathrm{dt}}=\frac{1}{6}(\mathrm{\Δ}{d}_{12}+12y_2)\·u_{\mathrm{dc}}+u_{\mathrm{Lz}12}---\left(19\right)$

Assuming that u _dckeep constant, Laplace transform done to above formula, can obtain:

$s(L_{m1}+L_{m2})I_{z2}=2U_{\mathrm{dc}}{Y}_2+\frac{\mathrm{\Δ}{D}_{12}}{6}{U}_{\mathrm{dc}}---\left(20\right)$

When the given electric current of each inverter is unequal or filter inductance is unequal, its zero axle all exists with d axle and q axle and is coupled, and coupling amount can produce interference to the control of zero axle, is the impact eliminating coupling amount, introduces the difference of two inverter non-zero duty ratios, i.e. Δ d ₁₂feedfoward control, the correction value y of zero vector can be obtained like this ₂:

$y_2=K_{p\_z}\·(i_{z2\_\mathrm{ref}}-i_{z2})+K_{i\_z}\∫(i_{z2\_\mathrm{ref}}-i_{z2})\mathrm{dt}-\frac{\mathrm{\Δ}{d}_{12}}{12}---\left(21\right)$

Obtain corresponding zero-sequence current ring control block diagram, as shown in Figure 8.

When the three pole reactor of two inverters all balances, even if when the given electric current of two inverters does not wait or its filter inductance is different, adopt and also can obtain good control effects in this way.But when three pole reactor change or the imbalance of two inverters, because the imbalance of inductance can produce Hz noise to zero axle, even if adopt the impact controlling in this way also to eliminate Hz noise, the control effects of circulation is poor.

Consider the uneven situation of each inverter three pole reactor, according to formula (13), in conjunction with the zero sequence Circulation Model of the shunt chopper under synchronous rotating frame, namely formula (4) can obtain:

$(L_{m1}+L_{m2})\frac{{\mathrm{di}}_{z2}}{\mathrm{dt}}=\frac{1}{6}(\mathrm{\Δ}{d}_{12}+12y_2)\·u_{\mathrm{dc}}+u_{\mathrm{Lz}12}---\left(22\right)$

Assuming that u _dckeep constant, Laplace transform done to above formula, can obtain:

$s(L_{m1}+L_{m2})I_{z2}=2U_{\mathrm{dc}}{Y}_2+\frac{\mathrm{\Δ}{D}_{12}}{6}{U}_{\mathrm{dc}}+U_{\mathrm{Lz}12}---\left(23\right)$

Δ D in above formula ₁₂, U _lz12be respectively variable Δ d ₁₂, u _lz12laplace transformation.

Visible, except zero vector correction value y ₂the impact that each inverter d axle and q shaft current controller export, the circulation of parallel system is also by the impact of the difference of two inverter inductor residual voltages, specifically, the missionary society of two inverter inductor residual voltages produces Hz noise to zero-axis current ring, make circulation contain certain power frequency composition, reduce the efficiency of system.In order to eliminate Hz noise, introduce the difference of two inverter inductor residual voltages at the basic above formula (21) of front a kind of control method, i.e. u _lz12uneoupled control, the correction value y of zero vector can be obtained like this ₂:

$y_2=K_{p\_z}\·(i_{z2\_\mathrm{ref}}-i_{z2})+K_{i\_z}\∫(i_{z2\_\mathrm{ref}}-i_{z2})\mathrm{dt}-\frac{\mathrm{\Δ}{d}_{12}}{12}-\frac{u_{\mathrm{Lz}12}}{2u_{\mathrm{dc}}}---\left(24\right)$

Obtain the control block diagram of zero-sequence current ring, as shown in Figure 9, like this, interference volume and feed-forward component are cancelled out each other, and after the impact of exclusive PCR amount, the control block diagram of parallel system zero-sequence current just can simplify, as shown in Figure 7.

Obtain the control block diagram of the whole system of two inverter parallel structures thus as shown in Figure 2, for any two inverters, because circulation equal and opposite in direction, the direction flow through are contrary, only need to control the circulation of one of them inverter, the circulation between two inverters can be controlled.First inverter 1 only controls d axle and q shaft current, and does not control zero-axis current, when carrying out SVPWM modulation, and zero vector V ₀and V ₇mean allocation.Second inverter 2, except controlling d axle and q shaft current, also will control zero-axis current.First to the zero-sequence current i of the second inverter 2 _z2sample; Then zero-sequence current controller is utilized to carry out PI adjustment to zero-sequence current, and introduce the feedfoward control of the difference of non-zero duty ratio and the difference of inductance residual voltage in two inverter SVPWM, the expression formula of zero-sequence current controller is as shown in formula (24); Zero-sequence current controller is finally utilized to export y ₂regulate in real time the distribution of zero vector in inverter 2SVPWM, the distribution of zero vector as shown in Figure 6.

The method also can be used for the individual inverter parallel structure of N (N >=3).For the individual inverter parallel structure of N (N >=3), wherein introduce the uneoupled control of inductance residual voltage at the circulation controller of any two inverters, the interference of inductance residual voltage to zero-axis current ring can be got rid of, obtain better circulation control effects.

Claims (5)

1. combining inverter parallel system circulation inhibition method under inductance unbalance condition, combining inverter in described combining inverter parallel system is common DC bus, the direct structure in parallel of AC, described parallel system adopts PI method to control, for the circulation inhibition method of above-mentioned combining inverter parallel system, the method is in two inverter parallel systems, the circulation of an inverter of combining inverter is controlled, specifically comprises the following steps:

Step one: to the zero-sequence current i of the second inverter (2) _z2sample, then perform step 2;

Step 2: utilize zero-sequence current PI controller to the zero-sequence current i of the second inverter (2) _z2with the zero-sequence current set-point i of the second inverter (2) _{z2_ref}do difference, and using the input signal of this difference as zero-sequence current PI controller, utilize the PI algorithm of zero-sequence current PI controller to regulate this input signal, obtain PI regulated value, then perform step 3 and step 4 simultaneously;

Step 3: by the difference △ d of the non-zero duty ratio of the space vector pulse width modulation of the first inverter (1) and the second inverter (2) ₁₂divided by 12, obtain the regulated quantity of non-zero, then perform step 5;

Step 4: by the difference u of the inductance residual voltage of the first inverter (1) and the second inverter (2) _lz12divided by the DC bus-bar voltage u of 2 times _dc, obtain the regulated quantity of inductance residual voltage, then perform step 5;

Step 5: the regulated quantity of the inductance residual voltage that the regulated quantity of the non-zero that the PI regulated value obtain step 2 and step 3 obtain and step 4 obtain is poor, the difference obtained is exported by zero sequence circulation controller 3, and this difference is as the correction value y of the second inverter (2) zero vector ₂, then perform step 6;

Step 6: the correction value y of the second inverter (2) zero vector utilizing step 5 to obtain ₂the distribution of zero vector in the space vector pulse width modulation of the second inverter (2) is regulated in real time, completes the suppression to circulation;

It is characterized in that, the correction value y of the second inverter (2) zero vector in step 5 ₂for:

$y_2=K_{p\_z}\·(i_{z2\_\mathrm{ref}}-i_{z2})+K_{i\_z}\∫(i_{z2\_\mathrm{ref}}-i_{z2})\mathrm{dt}-\frac{\mathrm{\Δ}{d}_{12}}{12}-\frac{u_{\mathrm{Lz}12}}{2u_{\mathrm{dc}}}---\left(4\right)$

K in formula _{p_z}for the proportionality coefficient of zero-sequence current PI controller, K _{i_z}for the integral coefficient of zero-sequence current PI controller.

2. combining inverter parallel system circulation inhibition method under inductance unbalance condition according to claim 1, is characterized in that, the zero-sequence current i of the second inverter (2) in step one _z2for:

$i_{z2}=\frac{i_{a2}+i_{b2}+i_{c2}}{3}---\left(1\right)$

I in formula _k2(k=a, b, c) is respectively a phase, b phase, the c phase current of the second inverter (2).

3. combining inverter parallel system circulation inhibition method under inductance unbalance condition according to claim 1, is characterized in that, the d of △ described in step 3 ₁₂for:

△d ₁₂＝(d ₂₁-d ₁₁)-(d ₂₂-d ₁₂) (2)

D in formula ₁₁be the duty ratio of first non-zero of the first inverter (1), d ₂₁be the duty ratio of second non-zero of the first inverter (1), d ₁₂be the duty ratio of first non-zero of the second inverter (2), d ₂₂it is the duty ratio of second non-zero of the second inverter (2).

4. combining inverter parallel system circulation inhibition method under inductance unbalance condition according to claim 1, is characterized in that, u described in step 4 _lz12for:

$\begin{array}{c}{u}_{L_{z12}}=u_{L_{z1}}-u_{L_{z2}}\\ =\frac{{\mathrm{\ωi}}_{d2}(L_{a2}\sin \mathrm{\ωt}+L_{b2}\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c2}\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}\\ -\frac{{\mathrm{\ωi}}_{d1}(L_{a1}\sin \mathrm{\ωt}+L_{b1}\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c1}\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}\\ +\frac{{\mathrm{\ωi}}_{q2}(L_{a2}\cos \mathrm{\ωt}+L_{b2}\cos (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c2}\cos (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}\\ -\frac{{\mathrm{\ωi}}_{q1}(L_{a1}\cos \mathrm{\ωt}+L_{b1}\cos (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})+L_{c1}\cos (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π}))}{3}\end{array}---\left(3\right)$

In formula be the inductance residual voltage of the first inverter (1), be the inductance residual voltage of the second inverter (2), i _dx(x=1,2) are respectively the d axle component of the three-phase current of the first inverter (1) and the second inverter (2), i _qx(x=1,2) are respectively the q axle component of the three-phase current of the first inverter (1) and the second inverter (2), L _kx(k=a, b, c; X=1,2) be respectively the inductance value of the first inverter (1) and second inverter a, b, c phase, ω is the angular frequency of electrical network.

5. combining inverter parallel system circulation inhibition method under inductance unbalance condition according to claim 1, is characterized in that, the concrete grammar of the real-time adjustment described in step 6 is:

In a control cycle T by the control of the on off state of the A phase of the second inverter (2) be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

In same control cycle T by the control of the on off state of the B phase of the second inverter (2) be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

In same control cycle T by the control of the on off state of the C phase of the second inverter (2) be: from the initial time of one-period, first continue low level time, then permanent High level time, finally continue low level time;

D ₁₁be the duty ratio of first non-zero of the first inverter (1), d ₂₁be the duty ratio of second non-zero of the first inverter (1), d ₁₂be the duty ratio of first non-zero of the second inverter (2), d ₂₂be the duty ratio of second non-zero of the second inverter (2), d ₀₂it is the duty ratio of the second inverter (2) zero vector.