CN104113077B - A kind of control method for coordinating of dual-feed asynchronous wind power generator high voltage crossing - Google Patents

Abstract

The invention discloses a kind of control method for coordinating of dual-feed asynchronous wind power generator high voltage crossing, its by dual-feed asynchronous wind power generator and net side and rotor-side converter Mathematical Modeling based on, during calculating line voltage high voltage, double-fed fan motor machine networking side and rotor-side converter are gained merit, the distribution principle of reactive power, give it to gain merit, the limit of reactive current is expressed, propose one based on this can effectively provide dynamic reactive support for electrical network and eliminate DC bus-bar voltage, active power, the high voltage crossing implementation of reactive power and electromagnetic torque fluctuation.The present invention can overcome the defects such as the weak and grid stability of Wind turbines high pressure ride-through capability in prior art difference, the target to generator optimal control when reaching line voltage high voltage; And can implement on original low voltage crossing hardware foundation completely, realize being connected with the existing low voltage crossing scheme of unit, form the broad sense voltage failure can tackling grid voltage amplitude cataclysm and pass through control strategy.

Description

A kind of control method for coordinating of dual-feed asynchronous wind power generator high voltage crossing

Technical field

The invention belongs to wind-driven generator control technology field, be specifically related to a kind of control method for coordinating of dual-feed asynchronous wind power generator high voltage crossing.

Background technology

In dual-feed asynchronous wind power generator (DFIG) system, generator unit stator is directly connected with electrical network by transformer, and the fault that electrical network occurs will directly have influence on the operation of generator itself.The finite capacity of its pwm converter be connected with rotor in addition, the control of limited capability only can be implemented to generator, compared with the wind generator system based on total power converter, double-fed asynchronous wind power system shows electric network fault very responsive, the feature that ability to bear is also poor.

At present, a large amount of invention and research are all about grid voltage sags fault on the impact of Wind turbines and low voltage crossing (LVRT) technology, and line voltage rise sharply on the research of the impact of Wind turbines and corresponding high voltage crossing (HVRT) technical field with invent relative less.Corresponding with Voltage Drop, voltage swells is also a kind of line voltage anomaly, occurs in line voltage and recovers or the power system reactive power superfluous moment.In the wind energy turbine set of configuration reactive power compensator; its reactive power is in dynamic equilibrium under normal circumstances; but when line voltage falls; often can cause some Wind turbines off-the-lines from electrical network without LVRT ability; now reactive power compensator is as regulated in time; power system reactive power will be caused superfluous, and then lifting the terminal voltage of site.As do not considered, overvoltage protection designs those units once tested by LVRT, now also has to continue off-the-line from electrical network, again causes the extensive off-grid of Wind turbines.Therefore, in the design of DFIG transient protection except need thinking better of Voltage Drop fault, also need the harm taking precautions against voltage swells, run with the high voltage crossing realizing unit.

Along with the continuous expansion of wind power generation installed capacity and the perfect of grid-connected criterion, having HVRT ability progressively becomes inevitable requirement to wind energy turbine set.Can be summarized as two large classes in recent years: based on the HVRT scheme increasing hardware circuit about the research of HVRT scheme and invention; Based on the HVRT scheme of improved system control strategy.Wherein hardware based HVRT scheme mainly implements the transformation of topological structure to promote Wind turbines HVRT ability by additional firmware equipment to double-fed wind energy converter.Obviously in these hardware based solutions, owing to adding complete hardware system, cost significantly increases, and design and the control of simultaneity factor are also more complicated.

And much based on the invention of the HVRT of improved system control strategy, do not thoroughly discuss the security of operation requirement of wind turbine networking side converter, rotor-side converter during line voltage rises sharply, two current transformers are not analyzed separately and mutual meritorious, reactive power restriction relation yet, the control strategy proposed not yet considers the dynamic reactive tenability of Wind turbines, be difficult to meet the networking requirement of grid-connected directive/guide to Wind turbines increasingly stringent, and some only gives the resolving ideas of HVRT from current transformer angle, do not take into full account the electro-magnetic transient characteristic of motor self.

Summary of the invention

For the above-mentioned technical problem existing for prior art, the invention provides a kind of control method for coordinating of dual-feed asynchronous wind power generator high voltage crossing, there is the advantage that Wind turbines high pressure ride-through capability is strong and grid adaptability is good, and to equipping Wind turbines the present invention of low voltage crossing facility without the need to increasing additional hardware, and control structure is simple, can reach good Static and dynamic performance.

A control method for coordinating for dual-feed asynchronous wind power generator high voltage crossing, as follows:

For the control of DFIG pusher side current transformer;

A1. the threephase stator voltage U of DFIG is gathered _sabc, threephase stator electric current I _sabc, three-phase rotor current I _rabc, rotational speed omega _rwith rotor position angle θ _r; Phase-locked loop (PLL) is utilized to extract the angular frequency of threephase stator voltage _sand phase theta _s, and then according to phase theta _sto threephase stator voltage U _sabc, threephase stator electric current I _sabcwith three-phase rotor current I _rabccarry out dq conversion, correspondence obtains the stator voltage vector U in d-q rotating coordinate system _sdq, stator current vector I _sdqwith rotor current vector I _rdq;

A2. according to rotor current vector I _rdqrotor voltage Front Feed Compensation Δ U is calculated by decoupling compensation algorithm _rdq; According to stator voltage vector U _sdqwith stator current vector I _sdqrotor current demagnetization compensation rate I is calculated by transient state flux compensation algorithm _{rdq_comp}, and the pusher side active-power P of DFIG is calculated by stator power _swith pusher side reactive power Q _s;

A3. make given pusher side with reference to active-power P _srefwith pusher side with reference to reactive power Q _srefdeduct pusher side active-power P respectively _swith pusher side reactive power Q _safter, regulated by PI and utilize rotor current demagnetization compensation rate I _{rdq_comp}compensate, obtain rotor current command I _{rdq_ref}; Make rotor current command I _{rdq_ref}deduct rotor current vector I _rdqafter, regulated by PI and utilize rotor voltage Front Feed Compensation Δ U _rdqcompensate, obtain rotor voltage instruction U _{rdq_ref};

A4. to rotor voltage instruction U _{rdq_ref}carry out Park inverse transformation obtain alpha-beta rest frame under rotor voltage instruction, and then by SVPWM (space vector pulse width modulation) structure obtain one group of switching signal to control DFIG pusher side current transformer;

For the control of DFIG net side converter;

B1. three-phase power grid voltage U is gathered _gabcwith three phase network electric current I _gabcand the DC bus-bar voltage V of DFIG net side converter _dcwith DC bus current I _dc; Phase-locked loop is utilized to extract the angular frequency of three-phase power grid voltage _gand phase theta _g, and then according to phase theta _gto three-phase power grid voltage U _gabcwith three phase network electric current I _gabccarry out dq conversion, correspondence obtains the line voltage vector U in d-q rotating coordinate system _gdqwith power network current vector I _gdq;

B2. according to line voltage vector U _gdqwith power network current vector I _gdqvoltage feed-forward control compensation rate Δ U is calculated by decoupling compensation algorithm _gdq; Make given reference DC bus-bar voltage V _dcrefdeduct DC bus-bar voltage V _dcregulated by PI afterwards, obtain netting side d shaft current command value I _gdref;

B3. net side d shaft current command value I is made _gdrefdeduct power network current vector I _gdqd axle component I _gdand add load-current feedforward compensation rate Δ I _load, make given net side q shaft current command value I _gqrefdeduct power network current vector I _gdqq axle component I _gq, and then regulated by PI and utilize voltage feed-forward control compensation rate Δ U _gdqcompensate, obtain voltage on line side instruction U _{gdq_ref};

B4. to voltage on line side instruction U _{gdq_ref}carry out Park inverse transformation obtain alpha-beta rest frame under voltage on line side instruction, and then by SVPWM structure obtain one group of switching signal to control DFIG net side converter.

Rotor current demagnetization compensation rate I is calculated by transient state flux compensation algorithm in described steps A 2 _{rdq_comp}detailed process as follows:

First, according to stator voltage vector U _sdqwith stator current vector I _sdq, the stator magnetic linkage ψ of DFIG is calculated by stator flux observer _s;

Then, stator magnetic linkage ψ is made _sbe separated by trapper and obtain its steady-state component ψ _sf, make stator magnetic linkage ψ _sdeduct its steady-state component ψ _sfobtain stator magnetic linkage ψ _stransient state component ψ _sn;

Finally, to transient state component ψ _snnamely rotor current demagnetization compensation rate I is obtained after carrying out scale amplifying _{rdq_comp}.

Described pusher side is with reference to active-power P _sreffor the maximal power point tracking mode instruction value of DFIG under normal grid conditions.

Described pusher side is with reference to reactive power Q _srefvalue under the condition meeting following relational expression:

${3U}_s(U_s-1.1)\≤Q_{\mathrm{sref}}\≤\frac{3}{2}{U}_s(\sqrt{{\left(\frac{L_m}{L_s}{I}_{r\max }\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}+\frac{U_s}{{\mathrm{\ω}}_1{L}_s})$

Wherein: U _sfor threephase stator voltage U _sabcamplitude, L _sfor the stator inductance of DFIG, L _mfor the rotor mutual inductance of DFIG, s is the slip of DFIG and s=(ω _s-ω _r)/ω _s, P _tfor the active power that DFIG is total, ω ₁for electrical network angular frequency, I _rmaxfor the maximum permissible current of DFIG pusher side current transformer.

Described net side q shaft current command value I _gqrefvalue under the condition meeting following relational expression:

I _gqref>=-2 (U _g-1.1) and

$-\sqrt{I_{g\max }^2-{(-\frac{2}{3}\frac{{\mathrm{sP}}_t}{(1-s)U_g})}^2}\≤I_{\mathrm{gqref}}\≤\frac{1}{{\mathrm{\ω}}_1{L}_g}(\sqrt{{V_{\mathrm{dc}}}^2/3-{\left({\mathrm{\ω}}_1{L}_g{I}_{\mathrm{gd}}\right)}^2}-U_g)$

Wherein: U _gfor three-phase power grid voltage U _gabcamplitude, I _gmaxfor the maximum permissible current of DFIG net side converter, s is the slip of DFIG and s=(ω _s-ω _r)/ω _s, P _tfor the active power that DFIG is total, ω ₁for electrical network angular frequency, L _gfor the inductance of reactor.

The present invention is based on the control method for coordinating of the double-fed wind power system high voltage crossing of reactive power support and demagnetization strategy, for the high voltage fail of middle generation at the scene in recent years, propose the Collaborative Control utilizing reactive power support and demagnetization strategy, for solving the problem of Wind turbines in high voltage fail; The method not only can realize the high voltage crossing with good dynamic; initiatively provide dynamic reactive support to faulty grids simultaneously; and then recover to provide contribution to faulty grids; and can implement on the hardware foundation of original low voltage crossing completely; without the need to increasing extra hardware protection device; thus can realize being connected with existing low voltage crossing scheme, form the broad sense voltage failure can tackling grid voltage amplitude cataclysm and pass through control strategy.

Therefore the present invention not only achieves stablizing in busbar voltage operation nargin, and eliminate DFIG system dc bus and fluctuation that is meritorious and reactive power in high voltage crossing process, during can not only guaranteeing line voltage high voltage, the not off-grid of double-fed fan motor unit runs, realize effectively controlling DFIG Wind turbines, certain dynamic reactive can also be provided to support to faulty grids, assist the fast quick-recovery of line voltage, the safe operation of other grid-connected loads favourable.

Accompanying drawing explanation

Fig. 1 is the structured flowchart of DFIG wind power system.

Fig. 2 is the flow chart of DFIG Wind turbines HVRT control method for coordinating of the present invention.

Fig. 3 is the concrete control block diagram of DFIG Wind turbines HVRT control method for coordinating of the present invention.

Fig. 4 is the simulation waveform figure of DFIG under Traditional control strategy; Wherein: u _gabcfor three-phase power grid voltage, i _gabcfor three phase network electric current, P _tand Q _tbe respectively the total active power of DFIG and reactive power, V _dcfor DC bus-bar voltage.

Fig. 5 is the simulation waveform figure only adopting dynamic reactive to support DFIG under control strategy; Wherein: U _{s (g)}for the amplitude of three-phase power grid voltage or threephase stator voltage, U _pccfor grid-connected point voltage, i _sabcfor threephase stator electric current, i _gabcfor three phase network electric current, P _sand Q _sbe respectively active power and the reactive power of DFIG pusher side, P _gand Q _gbe respectively active power and reactive power that DFIG nets side.

Fig. 6 is the simulation waveform figure of Stator transient magnetic linkage DC component attenuation process under coordination control strategy of the present invention; Wherein: ψ _sdand ψ _sqbe respectively d axle component and the q axle component of Stator transient magnetic linkage.

Fig. 7 is the simulation waveform figure of DFIG under coordination control strategy of the present invention; Wherein: T _efor electromagnetic torque, i _rabcfor three-phase rotor current.

Embodiment

In order to more specifically describe the present invention, below in conjunction with the drawings and the specific embodiments, technical scheme of the present invention is described in detail.

Mathematics model of stable state by synchronous rotating frame off line side PWM converter:

$\left\{\begin{array}{c}{U}_g=R_g{I}_{\mathrm{gd}}-{\mathrm{\ω}}_1{L}_g{I}_{\mathrm{gq}}+v_{\mathrm{gd}}\\ 0=R_g{I}_{\mathrm{gq}}+{\mathrm{\ω}}_1{L}_g{I}_{\mathrm{gd}}+v_{\mathrm{gq}}\end{array}\right.$

If R _g=0, then have:

$\left\{\begin{array}{c}{V}_{\mathrm{gd}}=U_g{+\mathrm{\ω}}_1{L}_g{I}_{\mathrm{gq}}\\ v_{\mathrm{gq}}={-\mathrm{\ω}}_1{L}_g{I}_{\mathrm{gd}}\end{array}\right.$

According to SVPWM principle, if do not produced ovennodulation, have:

$m=\frac{\sqrt{v_{\mathrm{gd}}^2+v_{\mathrm{gq}}^2}}{V_{\mathrm{dc}}/2}\≤\frac{\sqrt{3}}{2}$

Then can obtain a constraints of its reactive current during GSC (net side converter) normal work:

$I_{\mathrm{gq}}\≤\frac{1}{{\mathrm{\ω}}_1{L}_g}(\sqrt{{V_{\mathrm{dc}}}^2/3-{\left({\mathrm{\ω}}_1{L}_g{I}_{\mathrm{gd}}\right)}^2}-U_g)$

Figure 1 shows that DFIG system configuration and power flow distribution, then have:

$\left\{\begin{array}{c}{P}_t=P_s+P_g\\ Q_t=Q_s+Q_g\end{array}\right.$

Under ignoring DFIG stator and rotor winding copper loss and core loss condition, have again:

$\left\{\begin{array}{c}{P}_s=\frac{P_t}{1-s}\\ P_g=-\frac{s{P}_t}{1-s}\end{array}\right.$

In the synchronous rotating reference frame of d axle grid voltage orientation, the active power that net side PWM converter absorbs from electrical network and reactive power are respectively:

$\left\{\begin{array}{c}{P}_g=\frac{3}{2}{U}_g{I}_{\mathrm{gd}}\\ Q_g=-\frac{3}{2}{U}_g{I}_{\mathrm{gq}}\end{array}\right.$

Simultaneous can obtain:

$I_{\mathrm{gd}}=\frac{2}{3}\frac{P_g}{U_g}=-\frac{2}{3}\frac{s{P}_t}{(1-s)U_g}$

In order to not exceed the maximum permissible current I of GSC _gmax, the another one constraints of its reactive current when GSC normally works can be obtained:

$\left|I_{\mathrm{gq}}\right|\≤\sqrt{I_{g\max }^2-{I_{\mathrm{gd}}}^2}=\sqrt{I_{g\max }^2-{(-\frac{2}{3}\frac{s{P}_t}{(1-s)U_g})}^2}$

That is:

$-\sqrt{{\left(\frac{L_m}{L_s}{I}_{r\max }\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}-\frac{U_s}{{\mathrm{\ω}}_1{L}_s}\≤I_{\mathrm{sq}}\≤\sqrt{{\left(\frac{L_m}{L_s}{I}_{r\max }\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}-\frac{U_s}{{\mathrm{\ω}}_1{L}_s}$

The active current value of voltage high voltage period RSC is:

$I_{\mathrm{rd}}=-\frac{2}{3}\frac{P_s{L}_s}{U_s{L}_m}=-\frac{2}{3}\frac{L_s}{L_m}\frac{P_t}{(1-s)U_s}$

In order to not exceed the maximum permissible current I of RSC (pusher side converter) _rmax, have:

$\left|I_{\mathrm{rq}}\right|\≤\sqrt{I_{r\max }^2-{I_{\mathrm{rd}}}^2}$

RSC reactive current I _rqreactive current relation is exported with stator side:

$I_{\mathrm{rq}}=-\frac{L_s{I}_{\mathrm{sq}}}{L_m}-\frac{U_s}{{\mathrm{\ω}}_s{L}_m}$

Obtain the constraints of the reactive current of DFIG stator side:

$-\sqrt{{\left(\frac{L_m}{L_s}{I}_{r\max }\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}-\frac{U_s}{{\mathrm{\ω}}_1{L}_s}\≤I_{\mathrm{sq}}\≤\sqrt{{\left(\frac{L_m}{L_s}{I}_{r\max }\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}-\frac{U_s}{{\mathrm{\ω}}_1{L}_s}$

Fig. 2 is DFIG high-voltage ride through of wind power generating set control flow chart, when line voltage is lower than its normal float cap (present embodiment is set as 1.1 times of nominal values), GSC is operated in unity power factor control pattern, RSC is operated in maximal power point tracking pattern; Once detect that line voltage rises sharply to 1.1 times of nominal values and above, namely GSC wink switches to busbar voltage control model, and absorption portion reactive current; RSC then switches to reactive power support pattern.

Specifically, under normal circumstances, the current instruction value of GSC is as follows:

$\left\{\begin{array}{c}{I}_{\mathrm{gdref}}=\frac{2}{3}\frac{V_{\mathrm{dc}}}{U_g}[k_{\mathrm{vgp}}^{\′}(V_{\mathrm{dcref}}-V_{\mathrm{dc}})+k_{\mathrm{vgi}}^{\′}\∫(V_{\mathrm{dcref}}-V_{\mathrm{dc}})\mathrm{dt}]+\frac{2}{3}\frac{V_{\mathrm{dc}}}{U_g}{I}_{\mathrm{load}}\\ =k_{\mathrm{vgp}}(V_{\mathrm{dcref}}-V_{\mathrm{dc}})+k_{\mathrm{vgi}}\∫(V_{\mathrm{dcref}}-V_{\mathrm{dc}})\mathrm{dt}+\frac{2}{3}\frac{P_{\mathrm{load}}}{U_g}\\ I_{\mathrm{gqref}}=k_{\mathrm{qgp}}(Q_{\mathrm{gref}}-Q_g)+k_{\mathrm{qgi}}\∫(Q_{\mathrm{gref}}-Q_g)\mathrm{dt}\end{array}\right.$

Under normal circumstances, rotor-side current reference value is as follows:

$\left\{\begin{array}{c}{I}_{\mathrm{rdref}}=\frac{2}{3}\frac{L_s}{L_m}\frac{P_{\mathrm{sref}}}{U_s}\\ I_{\mathrm{rqref}}=\frac{2}{3}\frac{L_s}{L_m}\frac{Q_{\mathrm{sref}}}{U_s}-\frac{U_s}{{\mathrm{\ω}}_1{L}_m}\end{array}\right.$

When line voltage rises to more than 1.1 times of nominal values, the command value I of GSC active current _gdrefretentive control busbar voltage PI exports and remains unchanged, and its referenced reactive current value is as follows:

$\left\{\begin{array}{c}{I}_{\mathrm{gqref}}=-2(U_g-1.1)\frac{P_g}{P_s}\\ I_{\mathrm{gqref}\≥}-\sqrt{I_{g\max }^2-{(-\frac{2}{3}\frac{s{P}_t}{(1-s)U_g})}^2}\end{array}\right.$

When line voltage symmetry rises to more than 1.1 times of nominal values, RSC active current command value I _rdrefalso remain unchanged, its referenced reactive current value is as follows:

$\left\{\begin{array}{c}{I}_{\mathrm{sqref}}=-2(U_s-1.1)\\ Q_{\mathrm{sqref}}=-\frac{3}{2}{U}_s{I}_{\mathrm{sqref}}=3U_s(U_s-1.1)\\ Q_{\mathrm{sqref}}\≤\frac{3}{2}{U}_s(\sqrt{{\left(\frac{L_m}{L_s}{I}_{r\max }\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}+\frac{U_s}{{\mathrm{\ω}}_1{L}_s})\end{array}\right.$

When considering that the improvement based on rotor current ring during rotor demagnetization control strategy controls, the compensation term that above-mentioned active reactive current instruction value adds stator magnetic linkage fluctuation is as follows:

$\left\{\begin{array}{c}{i}_{\mathrm{rd}\_\mathrm{comp}}=-k_{\mathrm{damp}}{\mathrm{\ψ}}_{\mathrm{sdn}}\\ i_{\mathrm{rq}\_\mathrm{comp}}=-k_{\mathrm{damp}}{\mathrm{\ψ}}_{\mathrm{sqn}}\end{array}\right.$

By above-mentioned analysis, the control method for coordinating principle of present embodiment DFIG high voltage crossing as shown in Figure 3:

For the control of RSC;

A1. the threephase stator voltage U of DFIG is gathered _sabc, threephase stator electric current I _sabc, three-phase rotor current I _rabc, rotational speed omega _rwith rotor position angle θ _r; Phase-locked loop pll is utilized to extract the angular frequency of threephase stator voltage _sand phase theta _s, and then according to phase theta _sto threephase stator voltage U _sabc, threephase stator electric current I _sabcwith three-phase rotor current I _rabccarry out dq conversion, correspondence obtains the stator voltage vector U in d-q rotating coordinate system _sdq, stator current vector I _sdqwith rotor current vector I _rdq;

A2. according to rotor current vector I _rdqrotor voltage Front Feed Compensation Δ U is calculated by following decoupling compensation algorithm _rdq;

$\left\{\begin{array}{c}\mathrm{\Δ}{U}_{\mathrm{rd}}=-{\mathrm{\ω}}_{\mathrm{slip}}(-\frac{L_m}{L_s}{U}_s+\mathrm{\σ}{L}_r{I}_{\mathrm{rq}})\\ \mathrm{\Δ}{U}_{\mathrm{rq}}={\mathrm{\ω}}_{\mathrm{slip}}\mathrm{\σ}{L}_r{I}_{\mathrm{rd}}\end{array}\right.$

Wherein: U _sfor the amplitude of threephase stator voltage, ω _slipfor the slippage of DFIG and ω _slip=ω _s-ω _r, σ is the leakage inductance coefficient of DFIG and σ=1-(L _sl _r/ L _m ²), L _rfor the inductor rotor of DFIG, L _sfor the stator inductance of DFIG, L _mfor the rotor mutual inductance of DFIG;

According to stator voltage vector U _sdqwith stator current vector I _sdqrotor current demagnetization compensation rate I is calculated by following transient state flux compensation algorithm _{rdq_comp}:

First, stator flux observer is according to stator voltage vector U _sdqwith stator current vector I _sdqcalculate the stator magnetic linkage ψ of DFIG _s;

$\left\{\begin{array}{c}{\mathrm{\ψ}}_{\mathrm{sd}}=\∫(U_{\mathrm{sd}}-R_s{I}_{\mathrm{sd}})\mathrm{dt}\\ {\mathrm{\ψ}}_{\mathrm{sq}}=\∫(U_{\mathrm{sq}}-R_s{I}_{\mathrm{sq}})\mathrm{dt}\end{array}\right.$

Wherein: R _sfor the stator resistance of DFIG;

Then, stator magnetic linkage ψ is made _sbe separated by trapper and obtain its steady-state component ψ _sf, make stator magnetic linkage ψ _sdeduct its steady-state component ψ _sfobtain stator magnetic linkage ψ _stransient state component ψ _sn;

Finally, to DC component ψ _sn(proportionality coefficient is-K to carry out scale amplifying _damp) after namely obtain rotor current demagnetization compensation rate I _{rdq_comp}.

And then according to the pusher side active-power P of following formula DFIG _swith pusher side reactive power Q _s:

$\left\{\begin{array}{c}{P}_s=U_{\mathrm{sd}}{I}_{\mathrm{sd}}+U_{\mathrm{sq}}{I}_{\mathrm{sq}}\\ Q_s=U_{\mathrm{sq}}{I}_{\mathrm{sd}}-U_{\mathrm{sd}}{I}_{\mathrm{sq}}\end{array}\right.$

A3. make pusher side with reference to active-power P _sref(for the maximal power point tracking mode instruction value of DFIG when normal grid conditions) and pusher side are with reference to reactive power Q _srefdeduct pusher side active-power P respectively _swith pusher side reactive power Q _safter, regulated by PI and utilize rotor current demagnetization compensation rate I _{rdq_comp}compensate, obtain rotor current command I _{rdq_ref};

I _rdref＝I _rdcomp+PI(s)(P _sref-P _s)

I _rqref＝I _rqcomp+PI(s)(Q _sref-Q _s)

${3U}_s(U_s-1.1)\≤Q_{\mathrm{sref}}\≤\frac{3}{2}{U}_s(\sqrt{{\left(\frac{L_m}{L_s}{I}_{r\max }\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}+\frac{U_s}{{\mathrm{\ω}}_1{L}_s})$

Wherein: I _rdcompand I _rqcompbe respectively rotor current demagnetization compensation rate I _{rdq_comp}d axle component and q axle component, s is the slip of DFIG and s=(ω _s-ω _r)/ω _s, P _tfor the active power that DFIG is total, ω ₁for electrical network angular frequency, PI (s) is for PI adjustment is about the transfer function of Laplace domain; I _rmaxfor the maximum permissible current of RSC; Pusher side is with reference to reactive power Q _srefcan according to concrete electrical network code requirement value in above-mentioned scope.

Make rotor current command I _{rdq_ref}deduct rotor current vector I _rdqafter, regulated by PI and utilize rotor voltage Front Feed Compensation Δ U _rdqcompensate, obtain rotor voltage instruction U _{rdq_ref};

U _rdref＝ΔU _rd+PI(s)(I _rdref-I _rd)

U _rqref＝ΔU _rq+PI(s)(I _rqref-I _rq)

A4. to rotor voltage instruction U _{rdq_ref}carry out Park inverse transformation obtain alpha-beta rest frame under rotor voltage instruction, and then by SVPWM structure obtain one group of switching signal to control RSC.

For the control of GSC;

B1. three-phase power grid voltage U is gathered _gabcwith three phase network electric current I _gabcand the DC bus-bar voltage V of DFIG net side converter _dcwith DC bus current I _dc; Phase-locked loop pll is utilized to extract the angular frequency of three-phase power grid voltage _gand phase theta _g, and then according to phase theta _gcarry out dq conversion to three-phase power grid voltage and three phase network electric current, correspondence obtains the line voltage vector U in d-q rotating coordinate system _gdqwith power network current vector I _gdq;

B2. according to line voltage vector U _gdqwith power network current vector I _gdqvoltage feed-forward control compensation rate Δ U is calculated by following decoupling compensation algorithm _gdq;

$\left\{\begin{array}{c}{\mathrm{\ΔU}}_{\mathrm{gd}}={\mathrm{\ω}}_g{L}_g{I}_{\mathrm{gq}}+U_{\mathrm{gd}}\\ {\mathrm{\ΔU}}_{\mathrm{gq}}={\mathrm{\ω}}_g{L}_g{I}_{\mathrm{gd}}\end{array}\right.$

Wherein: L _gfor the inductance of reactor;

Make given reference DC bus-bar voltage V _dcrefdeduct DC bus-bar voltage V _dcregulated by PI afterwards, obtain netting side d shaft current command value I _gdre _f;

B3. net side d shaft current command value I is made _gdrefdeduct power network current vector I _gdqd axle component I _gdand add load-current feedforward compensation rate Δ I _load(Δ I _load=2V _dci _dc/ 3U _gd), make given net side q shaft current command value I _gqrefdeduct power network current vector I _gdqq axle component I _gq, and then regulated by PI and utilize voltage feed-forward control compensation rate Δ U _gdqcompensate, obtain voltage on line side instruction U _{gdq_ref};

U _gdref＝ΔU _gd+PI(s)(I _gdref-I _gd+ΔI _load)

U _gqref＝ΔU _gq+PI(s)(I _gqref-I _gq)

I _gqref>=-2 (U _g-1.1) and

$-\sqrt{I_{g\max }^2-{(-\frac{2}{3}\frac{{\mathrm{sP}}_t}{(1-s)U_g})}^2}\≤I_{\mathrm{gqref}}\≤\frac{1}{{\mathrm{\ω}}_1{L}_g}(\sqrt{{V_{\mathrm{dc}}}^2/3-{\left({\mathrm{\ω}}_1{L}_g{I}_{\mathrm{gd}}\right)}^2}-U_g)$

Wherein: U _gfor three-phase power grid voltage U _gabcamplitude, I _gmaxfor the maximum permissible current of GSC; Net side q shaft current command value I _gqrefcan according to concrete electrical network code requirement value in above-mentioned scope.

B4. to voltage on line side instruction U _{gdq_ref}carry out Park inverse transformation obtain alpha-beta rest frame under voltage on line side instruction, and then by SVPWM modulation technique structure obtain one group of pwm signal to control GSC.

With reference to above-mentioned control block diagram, net side, rotor-side converter are controlled, when unit set end voltage is more than 1.1 times of rated voltages, net side converter reactive current set-point and the motor stator side reactive power set-point of needs is calculated according to preceding method, and rotor demagnetization current offset values, and by these instructions Controling network side converter and rotor-side converter respectively.

Present embodiment passes through the output of net side converter and double-fed generator lagging reactive power, realizes the adjustment to grid-connected point voltage, inhibits the pump liter of DC bus-bar voltage simultaneously; By additional rotor demagnetization offset current, Stator transient magnetic linkage DC component is controlled on this basis, eliminate the fundamental frequency fluctuation of meritorious, reactive power and electromagnetic torque, improve performance and the effect of high voltage crossing.

Fig. 4 show line voltage rise sharply fault occur after, if GSC still adopts traditional unity power factor control strategy, certainly will cause DC bus-bar voltage be raised to exceed the maximum of DC bus can continued operation voltage 1200V, obviously this is that device institute is unallowed, and needs longer recovery time after being got rid of by causing trouble.Meanwhile, the instability of busbar voltage result also in the instability of grid side converter electric current, I _gabcharmonic content is large.The chain reaction that the work of GSC unsuccessfully must cause RSC to control, causes DFIG to have to from the off-the-line electrical network the most at last, HVRT failure.

Fig. 5 adopts reactive power support strategy of the present invention under showing identical grid conditions time, V _dcobtain effective control, be conducive to the fast quick-recovery of faulty grids and the safe operation of other grid-connected loads.Meanwhile, I _gabcharmonic content decreases, but because now i _gd ^*≠ 0, therefore I _gabclarger when comparatively running under unity power factor, but all the time at GSC capacity I _gmaxconstraint under.Although this control strategy restrained effectively the pump liter of DC bus-bar voltage, but the shake of uncontrollable DC bus-bar voltage, meritorious and reactive power.

Fig. 6 and Fig. 7 adopts the above-mentioned reactive power support strategy of the present invention and when rotor-side adopts demagnetization to control simultaneously under showing identical grid conditions, when getting a suitable K _dampduring value, Stator transient magnetic linkage decays to 0 very soon, thus eliminates the shake of DC bus-bar voltage, rotor-side electric current active reactive power and electromagnetic torque in a short period of time.The basis of previous step improves the dynamic property of system, makes DFIG unit successfully achieve high-performance high voltage crossing.

Claims (2)

1. a control method for coordinating for dual-feed asynchronous wind power generator high voltage crossing, as follows:

For the control of DFIG pusher side current transformer;

A1. the threephase stator voltage U of DFIG is gathered _sabc, threephase stator electric current I _sabc, three-phase rotor current I _rabc, rotational speed omega _rwith rotor position angle θ _r; Phase-locked loop is utilized to extract the angular frequency of threephase stator voltage _sand phase theta _s, and then according to phase theta _sto threephase stator voltage U _sabc, threephase stator electric current I _sabcwith three-phase rotor current I _rabccarry out dq conversion, correspondence obtains the stator voltage vector U in d-q rotating coordinate system _sdq, stator current vector I _sdqwith rotor current vector I _rdq;

A2. according to rotor current vector I _rdqrotor voltage Front Feed Compensation Δ U is calculated by decoupling compensation algorithm _rdq; According to stator voltage vector U _sdqwith stator current vector I _sdqrotor current demagnetization compensation rate I is calculated by transient state flux compensation algorithm _{rdq_comp}, and the pusher side active-power P of DFIG is calculated by stator power _swith pusher side reactive power Q _s;

Rotor current demagnetization compensation rate I is calculated by transient state flux compensation algorithm _{rdq_comp}detailed process as follows:

First, according to stator voltage vector U _sdqwith stator current vector I _sdq, the stator magnetic linkage ψ of DFIG is calculated by stator flux observer _s;

Then, stator magnetic linkage ψ is made _sbe separated by trapper and obtain its steady-state component ψ _sf, make stator magnetic linkage ψ _sdeduct its steady-state component ψ _sfobtain stator magnetic linkage ψ _stransient state component ψ _sn;

Finally, to transient state component ψ _snnamely rotor current demagnetization compensation rate I is obtained after carrying out scale amplifying _{rdq_comp};

A3. make given pusher side with reference to active-power P _srefwith pusher side with reference to reactive power Q _srefdeduct pusher side active-power P respectively _swith pusher side reactive power Q _safter, regulated by PI and utilize rotor current demagnetization compensation rate I _{rdq_comp}compensate, obtain rotor current command I _{rdq_ref}; Make rotor current command I _{rdq_ref}deduct rotor current vector I _rdqafter, regulated by PI and utilize rotor voltage Front Feed Compensation Δ U _rdqcompensate, obtain rotor voltage instruction U _{rdq_ref};

Described pusher side is with reference to reactive power Q _srefvalue under the condition meeting following relational expression:

$3U_s(U_s-1.1)\≤Q_{sref}\≤\frac{3}{2}{U}_s(\sqrt{{\left(\frac{L_m}{L_s}{I}_{rmax}\right)}^2-{(-\frac{2}{3}\frac{P_t}{(1-s)U_s})}^2}+\frac{U_s}{{\mathrm{\ω}}_1{L}_s})$

Wherein: U _sfor threephase stator voltage U _sabcamplitude, L _sfor the stator inductance of DFIG, L _mfor the rotor mutual inductance of DFIG, s is the slip of DFIG and s=(ω _s-ω _r)/ω _s, P _tfor the active power that DFIG is total, ω ₁for electrical network angular frequency, I _rmaxfor the maximum permissible current of DFIG pusher side current transformer;

A4. to rotor voltage instruction U _{rdq_ref}carry out Park inverse transformation obtain alpha-beta rest frame under rotor voltage instruction, and then by SVPWM structure obtain one group of switching signal to control DFIG pusher side current transformer;

For the control of DFIG net side converter;

B1. three-phase power grid voltage U is gathered _gabcwith three phase network electric current I _gabcand the DC bus-bar voltage V of DFIG net side converter _dcwith DC bus current I _dc; Phase-locked loop is utilized to extract the angular frequency of three-phase power grid voltage _gand phase theta _g, and then according to phase theta _gto three-phase power grid voltage U _gabcwith three phase network electric current I _gabccarry out dq conversion, correspondence obtains the line voltage vector U in d-q rotating coordinate system _gdqwith power network current vector I _gdq;

B2. according to line voltage vector U _gdqwith power network current vector I _gdqvoltage feed-forward control compensation rate Δ U is calculated by decoupling compensation algorithm _gdq; Make given reference DC bus-bar voltage V _dcrefdeduct DC bus-bar voltage V _dcregulated by PI afterwards, obtain netting side d shaft current command value I _gdref;

B3. net side d shaft current command value I is made _gdrefdeduct power network current vector I _gdqd axle component I _gdand add load-current feedforward compensation rate Δ I _load, make given net side q shaft current command value I _gqrefdeduct power network current vector I _gdqq axle component I _gq, and then regulated by PI and utilize voltage feed-forward control compensation rate Δ U _gdqcompensate, obtain voltage on line side instruction U _{gdq_ref};

Described net side q shaft current command value I _gqrefvalue under the condition meeting following relational expression:

I _gqref>=-2 (U _g-1.1) and

$-\sqrt{I_{gmax}^2-{(-\frac{2}{3}\frac{{\mathrm{sP}}_t}{(1-s)U_g})}^2}\≤I_{gqref}\≤\frac{1}{{\mathrm{\ω}}_1{L}_g}(\sqrt{{V_{dc}}^2/3-{\left({\mathrm{\ω}}_1{L}_g{I}_{gd}\right)}^2}-U_g)$

Wherein: U _gfor three-phase power grid voltage U _gabcamplitude, I _gmaxfor the maximum permissible current of DFIG net side converter, L _gfor the inductance of reactor;

B4. to voltage on line side instruction U _{gdq_ref}carry out Park inverse transformation obtain alpha-beta rest frame under voltage on line side instruction, and then by SVPWM structure obtain one group of switching signal to control DFIG net side converter.

2. control method for coordinating according to claim 1, is characterized in that: described pusher side is with reference to active-power P _sreffor the maximal power point tracking mode instruction value of DFIG under normal grid conditions.