CN101388639A - Non-position sensor vector control method for double-feed wind power generator - Google Patents

Abstract

The invention provides a method for controlling without-position sensor vector of a doubly-fed wind generator, wherein the control method adopts closed loop algorithm on the basis of a model reference self-adaptive module to indentify the rotor position and the rotary speed of a double fed wind induction generator. The control of the without-position sensor vector is possible, the cost of a generator system is lowered, and the anti-interference ability is improved. Simultaneously, the control method also adopts a method for improving voltage module observation flux linkages to observe a stator flux linkage under a stator static coordinate and a rotor flux linkage under a rotor static coordinate, the method can solve the problems that flux linkage phases which are obtained through observing the flux linkage through adopting a high pass filtering method are advanced and the amplitude is reduced, and the flux linkage which is corresponding with true value phases and amplitude is obtained.

Description

Non-position sensor vector control method for double-feed wind power generator

Technical field

The present invention relates to a kind of vector control method of double-fed wind power generator, particularly a kind of vector control method of implementing double-fed wind power generator under the condition of position-sensor-free belongs to technical field of wind power generation.

Background technology

In recent years, speed variant frequency constant dual feedback asynchronous wind generating technology is widely used.The double-fed wind power generator system is in operation, and needs motor speed and positional information accurately.Therefore, the position of high-res and velocity transducer are necessary.At present, the dual-feed asynchronous wind power generator group adopts the stator magnetic linkage oriented vector control technology of band position transducer to realize the decoupling zero control of the active power and the reactive power of electricity generation system mostly.

But the installation of speed and position transducer brings some new problems can for double-fed asynchronous wind generator system, reduces such as system reliability, and cost increases, and maintenance cost increases or the like.Yet in theory, people can by the voltage and current of double-fed wind power generator real-time calculate rotating speed of motor and rotor position angle.According to this theory, double-fed wind power generator does not need speed or position transducer still can realize the vector control of field orientation, i.e. position-sensor-free vector control.Double-fed wind power generator adopts the vector control method of this position-sensor-free, can save the line between generator and the current transformer, has strengthened the anti-interference and the reliability of control system, also will reduce hardware complexity, total cost and maintenance cost.The present invention promptly provides a kind of vector control method of implementing double-fed wind power generator under the condition of position-sensor-free.

Summary of the invention

The object of the present invention is to provide a kind of vector control method of under the condition of position-sensor-free, implementing double-fed wind power generator, by the enforcement of this method can reduce generator cost, reduce hardware complexity, improve the anti-interference and the reliability of system.

Goal of the invention of the present invention is achieved by following technical proposals:

Non-position sensor vector control method for double-feed wind power generator is characterized in that: comprise the steps:

(1) detects threephase stator voltage u _a, u _b, u _c, three-phase rotor current i _Ra, i _Rb, i _RcAnd DC bus-bar voltage u _Dc

(2) with described threephase stator voltage u _a, u _b, u _cObtain stator voltage u under the stator two-phase rest frame through 3/2 conversion _{S α}, u _{S β}Described three-phase rotor current i _Ra, i _Rb, i _RcObtain rotor current i under the rotor two-phase rest frame through 3/2 conversion _{R α}, i _{R β}

(3) with the stator voltage u under the described stator two-phase rest frame _{S α}, u _{S β}Process voltage model flux observation module obtains the stator magnetic linkage ψ under the stator rest frame _sWith stator magnetic linkage azimuth θ _sAgain with this stator magnetic linkage azimuth θ _sThrough differentiating, obtain synchronous angular velocity of rotation ω _s

(4) with the stator magnetic linkage ψ under the described stator rest frame _sWith stator magnetic linkage azimuth θ _sWith the rotor current i under the described rotor two-phase rest frame _{R α}, i _{R β}Calculate rotor position angle θ by the model reference adaptive module _rAnd rotational speed omega _r

(5) with described stator magnetic linkage azimuth θ _sWith rotor position angle θ _rSubtract each other the θ that obtains _s-θ _r, with θ _s-θ _rAs the angle of transformation of Park conversion, to the rotor current i under the described rotor two-phase rest frame _{R α}, i _{R β}Carry out the Park conversion, obtain the rotor current i under the synchronous rotating frame _Rd, i _Rq

(6) with the stator magnetic linkage ψ under the described stator rest frame _s, synchronous angular velocity of rotation ω _s, rotor speed ω _rAnd the rotor current i under the synchronous rotating frame _Rd, i _RqInput intersection amount computing module calculates calculating coupling terms u by this module _Rdc, u _Rqc, and meritorious idle reference quantity

(7) with described meritorious idle reference quantity

With the rotor current i under the described synchronous rotating frame _Rd, i _RqSubtract each other respectively, and after carrying out the PI conversion; Respectively with described coupling terms u _Rdc, u _RqcAddition obtains the rotor voltage reference value under the synchronous rotating frame

With this voltage reference value With θ _s-θ _rFor angle of transformation carries out anti-Park conversion, obtain the rotor voltage reference value under the rotor two-phase rest frame

(8) with described rotor voltage reference value

The input PWM generator, and according to described DC bus-bar voltage u _DcProduce the PWM control wave, drive the PWM current transformer.

The invention has the beneficial effects as follows:

1, in this position Sensorless Control algorithm, adopts rotor-position and rotating speed based on the closed loop algorithm identification dual-feed asynchronous wind power generator of model reference adaptive module.This makes the vector control of position-sensor-free become possibility, and makes the cost of generator system reduce, and its antijamming capability is improved.

2, in this position Sensorless Control algorithm, adopt the method for improving voltage model observation magnetic linkage that the rotor flux under stator magnetic linkage under the stator rest frame and the stationary rotor coordinate system is observed, this method can solve that to adopt high-pass filtering method to carry out the magnetic linkage phase place that flux observation obtains leading, the problem that amplitude diminishes obtains and actual value phase place and the consistent magnetic linkage of amplitude.

Description of drawings

Fig. 1 is position Sensorless Control algorithm principle figure;

Fig. 2 calculates the schematic diagram of rotor position angle and rotating speed for the model reference adaptive module;

Fig. 3 is the schematic diagram of the voltage model flux observation module after improving;

Fig. 4 is the schematic diagram of intersection amount computing module.

Embodiment

Below in conjunction with drawings and Examples the present invention is further described.

Double-fed generator is non-linear a, close coupling, multivariable system, and the conversion of its electromechanical energy is finished by fundamental wave magnetic field.Therefore, in order to simplify Mathematical Modeling, only consider the effect of double-fed generator air gap fundamental wave magnetic field usually, and ignore iron loss and ferromagnetic non-linear, the kelvin effect of winding and the temperature rise of rotor winding.Under synchronous rotating frame, when stator and all power taking motivations of rotor convention, the Mathematical Modeling of double-fed generator is in view of the above:

${\stackrel{\→}{u}}_s=R_s{\stackrel{\→}{i}}_s+p{\stackrel{\→}{\mathrm{\ψ}}}_s+j{\mathrm{\ω}}_s{\stackrel{\→}{\mathrm{\ψ}}}_s---\left(1\right)$

${\stackrel{\→}{u}}_r=R_r{\stackrel{\→}{i}}_r+p{\stackrel{\→}{\mathrm{\ψ}}}_r+j{\mathrm{\ω}}_{\mathrm{sl}}{\stackrel{\→}{\mathrm{\ψ}}}_r---\left(2\right)$

${\stackrel{\→}{\mathrm{\ψ}}}_s=L_s{\stackrel{\→}{i}}_s+{L_m\stackrel{\→}{i}}_r---\left(3\right)$

${\stackrel{\→}{\mathrm{\ψ}}}_r=L_m{\stackrel{\→}{i}}_s+{L_r\stackrel{\→}{i}}_r---\left(4\right)$

$T_e=-p_n(L_m/L_s)({\stackrel{\→}{\mathrm{\ψ}}}_s\⊗{\stackrel{\→}{i}}_s)---\left(5\right)$

Here,

Be respectively the rotor voltage under the synchronous rotating frame, stator and rotor current and rotor magnetic linkage.T _eBe electromagnetic torque, R _s, R _rIt is fixed rotor resistance; L _s, L _rIt is the rotor self-induction; L _mIt is mutual inductance; ω _s, ω _SlBe synchro angle frequency and slip-frequency; P _nBe the motor number of pole-pairs; P is a differential operator.

Consider when generator normally moves, stator winding directly links to each other with electrical network, and the generator unit stator frequency is constant on mains frequency, and the pressure drop on the stator resistance simultaneously also can be ignored for line voltage, under the condition, the stator magnetic linkage of generator is vertical with the stator voltage vector like this.Therefore, the d axle of synchronous rotating frame is placed on the stator magnetic linkage, controls the d of rotor, the decoupling zero that the q shaft current just can realize the active power and the reactive power of generator respectively.

According to the Mathematical Modeling of double-fed generator, rotating speed and position angle that we can derive rotor can be drawn as calculated by other measuring amount.Therefore, double-fed generator can be realized the vector control of position-sensor-free in theory.The present invention promptly is the vector control method according to a kind of position-sensor-free of this theoretical foundation design.Fig. 1 is position Sensorless Control algorithm principle figure of the present invention.As shown in the figure, non-position sensor vector control method for double-feed wind power generator of the present invention specifically comprises the steps:

(1) detects threephase stator voltage u _a, u _b, u _c, three-phase rotor current i _Ra, i _Rb, i _RcAnd DC bus-bar voltage u _Dc

(2) with the detected threephase stator voltage of step (1) u _a, u _b, u _cObtain stator voltage u under the stator two-phase rest frame through 3/2 conversion _{S α}, u _{S β}Detected three-phase rotor current i _Ra, i _Rb, i _RcObtain rotor current i under the rotor two-phase rest frame through 3/2 conversion _{R α}, i _{R β}

(3) the stator voltage u under the stator two-phase rest frame that step (2) is calculated _{S α}, u _{S β}Process voltage model flux observation module obtains the stator magnetic linkage ψ under the stator rest frame _sWith stator magnetic linkage azimuth θ _sBecause synchronous angular velocity of rotation ω _s=d θ _sSo/dt is again with resulting stator magnetic linkage azimuth θ _sThrough differentiating, can obtain synchronous angular velocity of rotation ω _s

(4) with the stator magnetic linkage ψ under the resulting stator rest frame in the step (3) _sWith stator magnetic linkage azimuth θ _sWith the rotor current i under the resulting rotor two-phase of step (2) rest frame _{R α}, i _{R β}By the closed loop algorithm of model reference adaptive module, identification obtains rotor position angle θ _rAnd rotational speed omega _r

(5) the stator magnetic linkage azimuth θ that step (3) is calculated _sThe rotor position angle θ that calculates with step (4) _rSubtract each other the θ that obtains _s-θ _r, with θ _s-θ _rAs the angle of transformation of Park conversion, the rotor current i under the rotor two-phase rest frame that step (2) is calculated _{R α}, i _{R β}Carry out the Park conversion, obtain the rotor current i under the synchronous rotating frame _Rd, i _Rq

(6) the stator magnetic linkage ψ under the stator rest frame that step (3) is obtained _sWith synchronous angular velocity of rotation ω _s, the rotational speed omega that step (4) obtains _rAnd the rotor current i under the synchronous rotating frame that calculates of step (5) _Rd, i _RqInput intersection amount computing module calculates calculating coupling terms u by this module _Rdc, u _Rqc, and given meritorious idle reference quantity

(7) the meritorious idle reference quantity that step (6) is calculated

Rotor current i under the synchronous rotating frame that calculates with step (5) _Rd, i _RqSubtract each other respectively, and after carrying out the PI conversion, respectively the coupling terms u that calculates with step (6) _Rdc, u _RqcAddition obtains the rotor voltage reference value under the synchronous rotating frame

Should

With resulting θ in the step (5) _s-θ _rCarry out anti-Park conversion as angle of transformation, obtain the rotor voltage reference value under the rotor two-phase rest frame

(8) with the rotor voltage reference value that obtains in the step (7)

The input PWM generator, and according to detected DC bus-bar voltage u in the step (1) _DcProduce the PWM control wave, drive the PWM current transformer.

Wherein, the model reference adaptive module that adopts in the step (4) is by stator magnetic linkage ψ _s, stator magnetic linkage azimuth θ _s, rotor current i _{R α}, i _{R β}Calculate rotor position angle θ _rAnd rotational speed omega _r, be that the present invention is able to the key modules implemented under the condition of position-sensor-free.The main thought of this model reference adaptive module controls is to remain the equation of estimated parameter as adjustable model with containing, and as the reference model, two models have the output variable of same physical meaning with the equation that do not contain unknown parameter.Two models are worked simultaneously, and the difference of utilizing its output variable adjusts the parameter of adjustable model in real time according to suitable adaptive rate, with the purpose of the output tracking reference model that reaches controlling object.

In the present invention, the voltage model of selecting to mention in the step (3) is as the reference model, other models that will contain rotor-position or rotary speed information promptly can use the rotor-position and the rotating speed of model reference adaptive modular approach identification double-fed wind power generator as adjustable model.Fig. 2 promptly is the schematic diagram that this model reference adaptive module is calculated rotor position angle and rotating speed.Should illustrate, owing in this link, relate to the conversion between stator rest frame and the stationary rotor coordinate system.Below in order to distinguish each amount, on the mark of parameter, be illustrated in vector in the stator rest frame with subscript " s ", be illustrated in vector in the stationary rotor coordinate system with subscript " r ".As shown in the figure, described step (4) realizes by following concrete steps:

(4A) the rotor current i under the rotor two-phase rest frame that step (2) is obtained _{R α}, i _{R β}After the PI conversion, obtain the rotor voltage u under the two-phase rest frame _{R α}, u _{R β}

(4B) with described rotor voltage u _{R α}, u _{R β}Input voltage model flux observation module obtains the rotor flux under the stationary rotor coordinate system

Here, the voltage model flux observation modular structure of being carried in described voltage model flux observation module and the abovementioned steps (3) is identical, all is the observation modules that obtain associated voltage model magnetic linkage by the voltage under the two-phase rest frame.

(4C) the rotor current i under the rotor two-phase rest frame that obtains of step (2) _{R α}, i _{R β}After polar coordinate transform, obtain the rotor current vector under the stationary rotor coordinate system through rectangular coordinate

(4D) with above-mentioned gained rotor current vector

With the rotor flux that calculates in the step (4B)

Pass through calculating formula ${\stackrel{\→}{\mathrm{\ψ}}}_s^r=\frac{L_s}{L_m}{\stackrel{\→}{\mathrm{\ψ}}}_r^r-\frac{L_s{L}_r-L_m^2}{L_m}{\stackrel{\→}{i}}_r^r$ Calculate the stator magnetic linkage under the stationary rotor coordinate system

Pass through rotor position angle then

With this stator magnetic linkage

To the stator rest frame, obtain stator magnetic linkage vector under the adjustable stator rest frame by the stationary rotor coordinate system transformation ${\stackrel{\stackrel{\→}{^}}{\mathrm{\ψ}}}_s^s={\stackrel{\→}{\mathrm{\ψ}}}_s^r{e}^{j{\widehat{\mathrm{\θ}}}_r}.$

Here, described calculating formula ${\stackrel{\→}{\mathrm{\ψ}}}_s^r=\frac{L_s}{L_m}{\stackrel{\→}{\mathrm{\ψ}}}_r^r-\frac{L_s{L}_r-L_m^2}{L_m}{\stackrel{\→}{i}}_r^r,$ Shift onto and draw by aforementioned formula (3), formula (4).Obtain by described formula (4) conversion ${\stackrel{\→}{i}}_s=({\stackrel{\→}{\mathrm{\ψ}}}_r^r-L_r{\stackrel{\→}{i}}_r^r)/L_m.$ Substitution formula (3) obtains: ${\stackrel{\→}{\mathrm{\ψ}}}_s^r=\frac{L_s}{L_m}{\stackrel{\→}{\mathrm{\ψ}}}_r^r-\frac{L_s{L}_r-L_m^2}{L_m}{\stackrel{\→}{i}}_r^r.$

Here, because the stator magnetic linkage that draws all is based on the vector calculating stationary rotor coordinate system under in the adjustable model, thus to calculate used vector and all added subscript " r " in order to distinguish mutually with the stator magnetic linkage of reference model, promptly

(4E) the stator magnetic linkage vector under the stator rest frame that obtains in the calculation procedure (3)

And stator magnetic linkage vector under the adjustable stator rest frame that obtains in the step (4D) ${\stackrel{\stackrel{\→}{^}}{\mathrm{\ψ}}}_s^s={\stackrel{\→}{\mathrm{\ψ}}}_s^r{e}^{j{\widehat{\mathrm{\θ}}}_r}$ Cross product $e={\stackrel{\stackrel{\→}{^}}{\mathrm{\ψ}}}_s^s\⊗{\stackrel{\→}{\mathrm{\ψ}}}_s^s={\stackrel{\→}{\mathrm{\ψ}}}_{\mathrm{s\α}}{\mathrm{\ψ}}_{\mathrm{s\β}}-{\widehat{\mathrm{\ψ}}}_{\mathrm{s\β}}{\mathrm{\ψ}}_{\mathrm{s\α}};$

(4F) cross product that step (4E) is calculated as a result e carry out obtaining adjustable rotating speed after the PI conversion Then to this adjustable rotating speed

Obtain adjustable rotor position angle behind the integration

(4G) should adjustable rotor position angle

As the angle of transformation of adjustable model, repeating step (4C) to (4F) stator magnetic linkage vector under the stator rest frame of adjustable model output

Stator magnetic linkage vector under the stator rest frame of reference model output in the tracking

Till, promptly judge

Whether stable, if stable, export above-mentioned adjustable rotating speed

With adjustable rotor position angle

The adjustable rotating speed of this moment

Be rotor speed ω _r, adjustable rotor position angle

Be rotor position angle θ _r

By above-mentioned steps as can be known, the stator magnetic linkage vector of model reference adaptive module of the present invention to calculate through step (3)

Be reference model, to contain rotor position angle θ _rThe stator magnetic linkage vector be adjustable model.By regulating rotor position angle θ _rMake adjustable model approach reference model, till following the tracks of.When adjustable model during with reference model to sum up, we can think rotor position angle θ at this moment _rBe true value.

We are not difficult to find out by above-mentioned steps, and the reference model that is relied in model reference adaptive module computational process is to calculate by the voltage model flux observation module in the abovementioned steps (3) to get; And the calculating of adjustable model also depends on voltage model flux observation module.As seen, the accuracy of the accuracy of this model reference adaptive module calculating and the calculating of voltage model flux observation module has close getting in touch.

Yet the flux observation module of existing voltage model is to adopt the method for stator back electromotive force integration, promptly $\stackrel{\→}{\mathrm{\ψ}}=\∫(\stackrel{\→}{u}-R\stackrel{\→}{i})\mathrm{dt}.$ This method is fairly simple, but because it without any correction mechanism, causes working as voltage

Electric current

When producing the measure error or the error of calculation, will produce a drift by integration with stator resistance R.Particularly when the electrical network three-phase voltage falls suddenly, adopt the method observation magnetic linkage of this pure integration will produce more serious integrator drift problem.

In order to eliminate this drift value, the general method that adopts is to allow integral result pass through a high pass filter.Because the transfer function of integral element is 1/s, the transfer function of high pass filter is s/ (s+ ω _c) (ω wherein _cBe cut-off frequency), so adopt this integral result to represent to be with transfer function by the method for high pass filter $\stackrel{\→}{\mathrm{\ψ}}=(\stackrel{\→}{u}-R\stackrel{\→}{i})/(s+{\mathrm{\ω}}_c).$

Adopt this method, the integrator drift amount is because frequency is 0, when the high pass filter by filtering.But, from transfer function $\stackrel{\→}{\mathrm{\ψ}}=(\stackrel{\→}{u}-R\stackrel{\→}{i})/(s+{\mathrm{\ω}}_c)$ As can be seen, it is an integral element

With high-pass filtering link s/ (s+ ω _c) product.Because ω here _cBe one greater than 0 constant, for any real part greater than for 0 the s, | s/ (s+ ω _c) | all less than 1, will cause that like this magnetic linkage that observes is littler than actual magnetic linkage amplitude; Analyze s/ (s+ ω simultaneously _c) phase-frequency characteristic, its phase-frequency characteristic is arctan (ω _c/ ω), promptly to any greater than 0 ω, the phase angle of this high-pass filtering link is all between 0～pi/2.So after this link introducing, can cause that the magnetic linkage phase place that observes is leading.At above problem, the present invention also provides the improvement project of this voltage model flux observation module, to solve the problems referred to above that exist in the existing voltage model flux observation module.

Fig. 3 is the schematic diagram of the voltage model flux observation module after improving.In this flux observation method, integral element and low pass filter are combined the observation of carrying out stator magnetic linkage, concrete:

$y=\frac{1}{s+{\mathrm{\ω}}_c}x+\frac{{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}z---\left(25\right)$

Y is integral element output in the formula (25), and x is the input of integral element, and z is a compensating signal, ω _cIt is cut-off frequency.Can find that from formula (25) z gets 0 or during y, formula (25) can be considered as high pass link and pure integral element respectively respectively.Suitable adjustment compensation rate can make formula (25) play effect between pure integral element and first order inertial loop.

When as the motor flux observation, y is a magnetic linkage

Because

With respect to voltage

Very little, can ignore, so, when the observation magnetic linkage, can directly use voltage

Replace So x is a voltage

When z got 0, formula (25) became $\stackrel{\→}{\mathrm{\ψ}}=\stackrel{\→}{u}/(s+{\mathrm{\ω}}_c),$ Be pure integral element and multiply by the computational methods of high-pass filtering link.When z got y, formula (25) became $\stackrel{\→}{\mathrm{\ψ}}=\stackrel{\→}{u}/(s+{\mathrm{\ω}}_c)+{\mathrm{\ω}}_c\stackrel{\→}{\mathrm{\ψ}}/(s+{\mathrm{\ω}}_c),$ So $\stackrel{\→}{\mathrm{\ψ}}=\stackrel{\→}{u}/s$ Be pure integral element.So, when z gets the suitable value that levels off to y, just can make detected magnetic linkage level off to the theoretical value of pure integral element $\stackrel{\→}{\mathrm{\ψ}}=\stackrel{\→}{u}/s,$ Diminish leading with regard to the unlikely amplitude of observation magnetic linkage that causes like this with phase place; Simultaneously, because in computational process, comprise the high-pass filtering link, can eliminate the existing drift value of pure integral element observation magnetic linkage.

The concrete grammar of the voltage model flux observation module after this improvement comprises following steps:

(3A) the voltage u under the two-phase rest frame _α, u _βCarry out high-pass filtering behind the integration respectively, obtain the ψ after the high-pass filtering _{α h}, ψ _{β h}:

${\mathrm{\ψ}}_{\mathrm{\αh}}=\frac{u_{\mathrm{\α}}}{s+{\mathrm{\ω}}_c}$

${\mathrm{\ψ}}_{\mathrm{\βh}}=\frac{u_{\mathrm{\β}}}{s+{\mathrm{\ω}}_c}$

Wherein ${\mathrm{\ω}}_c=\frac{1}{\mathrm{\τ}},$ τ is a time constant filter;

(3B) ψ that the last cycle is calculated _α, ψ _βCarry out rectangular coordinate to polar coordinate transform, obtain the flux linkage vector amplitude ψ under the polar coordinate system _cAnd angular position theta _c

(3C) to flux linkage vector amplitude ψ _cAfter carrying out amplitude limiting processing, with the angular position theta that obtains in the step (3B) _cCarry out the coordinate transform of polar coordinates as the position angle of new flux linkage vector, obtain the magnetic linkage feedback quantity ψ under the two-phase rest frame to rectangular coordinate _{α c}, ψ _{β c}

Here to flux linkage vector amplitude ψ _cCarrying out amplitude limiting processing, is that when reaching capacity state, magnetic linkage can not continue to increase, so in calculating, consider this situation, it is carried out amplitude limiting processing because ferromagnetic saturated because of existing in the real electrical machinery.

(3D) the magnetic linkage feedback quantity ψ that obtains of step (3C) _{α c}, ψ _{β c}After carrying out low-pass filtering respectively, obtain the ψ after the low-pass filtering _{α l}, ψ _{β l}:

${\mathrm{\ψ}}_{\mathrm{\αl}}=\frac{{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}{\mathrm{\ψ}}_{\mathrm{\αc}}$

${\mathrm{\ψ}}_{\mathrm{\βl}}=\frac{{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}{\mathrm{\ψ}}_{\mathrm{\βc}}$

(3E) ψ that obtains of step (3A) _{α h}, ψ _{β h}The ψ that obtains with step (3D) _{α l}, ψ _{β l}After the addition, obtain the magnetic linkage ψ in this cycle respectively _α, ψ _β

(3F) the magnetic linkage ψ that obtains by step (3E) _α, ψ _βFormula as calculated $\mathrm{\ψ}=\sqrt{{\mathrm{\ψ}}_{\mathrm{\α}}^2+{\mathrm{\ψ}}_{\mathrm{\β}}^2}$ Calculate the magnetic linkage amplitude ψ under the rest frame; Magnetic linkage ψ _α, ψ _βFormula as calculated $\mathrm{\θ}=\arctan \frac{{\mathrm{\ψ}}_{\mathrm{\β}}}{{\mathrm{\ψ}}_{\mathrm{\α}}}$ Calculate the flux linkage vector angular position theta; With the magnetic linkage amplitude ψ that calculated and flux linkage vector angular position theta formula as calculated $\stackrel{\→}{\mathrm{\ψ}}=\mathrm{\ψ}{e}^{\mathrm{j\θ}}$ Calculate the flux linkage vector under the rest frame

As from the foregoing, adopt the voltage model flux observation module after this improvement to represent with following formula:

${\mathrm{\ψ}}_{\mathrm{\α}}\left(n\right)=\frac{u_{\mathrm{\α}}}{s+{\mathrm{\ω}}_c}+\frac{{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}{\mathrm{\ψ}}_{\mathrm{\α}}(n-1)$

${\mathrm{\ψ}}_{\mathrm{\β}}\left(n\right)=\frac{u_{\mathrm{\β}}}{s+{\mathrm{\ω}}_c}+\frac{{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}{\mathrm{\ψ}}_{\mathrm{\β}}(n-1)$

ψ wherein _α(n), ψ _β(n) be magnetic linkage under this cycle rest frame, ψ _α(n-1), ψ _β(n-1) be magnetic linkage under the rest frame of last cycle; Because the control cycle of motor is very short, in this case, can suppose ψ _α(n) ≈ ψ _α(n-1), ψ _β(n) ≈ ψ _β(n-1), therefore:

${\mathrm{\ψ}}_{\mathrm{\α}}\left(n\right)=\frac{u_{\mathrm{\α}}}{s}\·\frac{s}{s+{\mathrm{\ω}}_c}+\frac{{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}{\·\mathrm{\ψ}}_{\mathrm{\α}}(n-1)\≈{\mathrm{\ψ}}_{\mathrm{\α}}\left(n\right)\·\frac{s+{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}={\mathrm{\ψ}}_{\mathrm{\α}}\left(n\right)$

${\mathrm{\ψ}}_{\mathrm{\β}}\left(n\right)=\frac{u_{\mathrm{\β}}}{s}\·\frac{s}{s+{\mathrm{\ω}}_c}+\frac{{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}{\·\mathrm{\ψ}}_{\mathrm{\β}}(n-1)\≈{\mathrm{\ψ}}_{\mathrm{\β}}\left(n\right)\·\frac{s+{\mathrm{\ω}}_c}{s+{\mathrm{\ω}}_c}={\mathrm{\ψ}}_{\mathrm{\β}}\left(n\right)$

So, adopt the voltage model flux observation modular approach after this improvement can observe the magnetic linkage identical theoretically with really being magnetic linkage amplitude and phase place.

In stator magnetic linkage oriented double-fed asynchronous wind generator system, under voltage vector continually varying situation, magnetic linkage also changes continuously, and in this case, the rotor flux under stator magnetic linkage under the stator rest frame and the stationary rotor coordinate system all satisfies ψ _α(n) ≈ ψ _α(n-1), ψ _β(n) ≈ ψ _β(n-1), so, adopt the voltage model flux observation module after this improvement can obtain stator magnetic linkage and rotor flux accurately.

Fig. 4 is the schematic diagram of intersection amount computing module in the step (6), and this module is in order to the cross reference amount in the compute vectors control procedure.

During the operation of double-fed generator nominal situation, the pressure drop on the stator winding is very little for line voltage, can ignore.If stator magnetic linkage ψ _sKeep constant, the fundamental equation of the double-fed generator of rewriting formula (1)～formula (5):

u _sd＝R _si _sd+pψ _sd-ω _sψ _sq (6)

u _sq＝R _si _sq+pψ _sq+ω _sψ _sd (7)

u _rd＝R _ri _rd+pψ _rd-ω _slψ _rq (8)

u _rq＝R _ri _rq+pψ _rq+ω _slψ _rd (9)

ψ _sd＝L _si _sd+L _mi _rd (10)

ψ _sq＝L _si _sq+L _mi _rq (11)

ψ _rd＝L _mi _sd+L _ri _rd (12)

ψ _rq＝L _mi _sq+L _ri _rq (13)

T _e＝p _nL _m(i _sqi _rd-i _sdi _rq) (14)

When being oriented in the d axle of synchronous rotating frame on the stator magnetic linkage, the q axle component ψ of stator magnetic linkage _SqBe 0, then above-mentioned formula (the 10)～magnetic linkage of formula (14) and the expression formula of electromagnetic torque can be reduced to:

ψ _sd＝L _si _sd+L _mi _rd＝L _mi _ms＝ψ _s (15)

$i_{\mathrm{sq}}=-\frac{L_m}{L_s}{i}_{\mathrm{rq}}---\left(16\right)$

${\mathrm{\ψ}}_{\mathrm{rd}}=\frac{L_m^2}{L_s}{i}_{\mathrm{ms}}+{\mathrm{\σL}}_r{i}_{\mathrm{rd}}---\left(17\right)$

ψ _rq＝σL _ri _rq (18)

$T_e=-P_{\mathrm{<figure-callout\; id="2"\; label="n\; L\; m"\; filenames="A200810225848E00181.png,A200810225848E00192.png"\; state="\{\{state\}\}">n</figure-callout>}}{L}_m^{}{}_2^{}/L_r{i}_{\mathrm{ms}}{i}_{\mathrm{rq}}---\left(19\right)$

In the formula, u here _Sd, u _Sq, u _Rd, u _Rq, i _Sd, i _Sq, i _Rd, i _Rq, ψ _Sd, ψ _Sq, ψ _Rd, ψ _RqBe respectively the rotor voltage under the synchronous rotating frame, stator and rotor current and rotor magnetic linkage.T _eBe electromagnetic torque, R _s, R _rIt is fixed rotor resistance; L _s, L _rIt is the rotor self-induction; L _mIt is mutual inductance; ω _s, ω _SlBe synchro angle frequency and slip-frequency; P _nBe the motor number of pole-pairs; P is a differential operator; i _MsBe the broad sense exciting current; $\mathrm{\&sigma;}=1-{\mathrm{<figure-callout\; id="2"\; label="L\; m"\; filenames="A200810225848E00181.png,A200810225848E00192.png"\; state="\{\{state\}\}">L</figure-callout>}}_m^{}/L_s{l}_r$ Be magnetic leakage factor.

By formula (6) as can be known, the d axle component of stator voltage vector approaches 0, and at this moment, stator magnetic linkage is approximate vertical mutually with the stator voltage vector.There is following relational expression in stable state:

${\mathrm{\&psi;}}_s={\mathrm{\&psi;}}_{\mathrm{sd}}\&ap;\frac{u_{\mathrm{sq}}}{{\mathrm{\&omega;}}_s}---\left(20\right)$

By formula (20) as can be known, when keeping line voltage amplitude and frequency constant, stator magnetic linkage is constant, stator broad sense exciting current i _MsSize is constant.

Formula (17), formula (18) difference substitution formula (8), formula (9) can be obtained:

$u_{\mathrm{rd}}^{\mathrm{ref}}=R_r{i}_{\mathrm{rd}}+{\mathrm{\&sigma;L}}_r\frac{{\mathrm{di}}_{\mathrm{rd}}}{\mathrm{dt}}-{\mathrm{\&omega;}}_{\mathrm{sl}}\mathrm{\&sigma;}{L}_r{i}_{\mathrm{rq}}---\left(21\right)$

$u_{\mathrm{rd}}^{\mathrm{ref}}=R_r{i}_{\mathrm{rq}}+{\mathrm{\&sigma;L}}_r\frac{{\mathrm{di}}_{\mathrm{rq}}}{\mathrm{dt}}+{\mathrm{\&omega;}}_{\mathrm{sl}}\frac{L_m^2}{L_s}{i}_{\mathrm{ms}}+{\mathrm{\&omega;}}_{\mathrm{sl}}\mathrm{\&sigma;}{L}_r{i}_{\mathrm{rd}}---\left(22\right)$

In the formula,

It is respectively the rotor voltage reference value.The coupling terms u that definition is caused by back electromotive force _Rdc, u _RqcAs follows:

u _rdc＝-ω _slσL _ri _rq (23)

$u_{\mathrm{rqc}}={\mathrm{\&omega;}}_{\mathrm{sl}}\frac{L_m^2}{L_s}{i}_{\mathrm{ms}}+{\mathrm{\&omega;}}_{\mathrm{sl}}\mathrm{\&sigma;}{L}_r{i}_{\mathrm{rd}}---\left(24\right)$

In sum, in the designed position Sensorless Control algorithm of the present invention, adopt the method for improving voltage model observation magnetic linkage that the rotor flux under stator magnetic linkage under the stator rest frame and the stationary rotor coordinate system is observed, this method can solve that to adopt high-pass filtering method to carry out the magnetic linkage phase place that flux observation obtains leading, the problem that amplitude diminishes obtains and actual value phase place and the consistent magnetic linkage of amplitude; Employing is based on the rotor-position and the rotating speed of the closed loop algorithm identification dual-feed asynchronous wind power generator of model reference adaptive module.This method has good dynamic and steady-state behaviour.

Claims (4)

1, non-position sensor vector control method for double-feed wind power generator is characterized in that: comprise the steps:

(1) detects threephase stator voltage u _a, u _b, u _c, three-phase rotor current i _Ra, i _Rb, i _RcAnd DC bus-bar voltage u _Dc

(2) with described threephase stator voltage u _a, u _b, u _cObtain stator voltage u under the stator two-phase rest frame through 3/2 conversion _{S α}, u _{S β}Described three-phase rotor current i _Ra, i _Rb, i _RcObtain rotor current i under the rotor two-phase rest frame through 3/2 conversion _{R α}, i _{R β}

(3) with the stator voltage u under the described stator two-phase rest frame _{S α}, u _{S β}Process voltage model flux observation module obtains the stator magnetic linkage ψ under the stator rest frame _sWith stator magnetic linkage azimuth θ _sAgain with this stator magnetic linkage azimuth θ _sThrough differentiating, obtain synchronous angular velocity of rotation ω _s

(4) with the stator magnetic linkage ψ under the described stator rest frame _sWith stator magnetic linkage azimuth θ _sWith the rotor current i under the described rotor two-phase rest frame _{R α}, i _{R β}Calculate rotor position angle θ by the model reference adaptive module _rAnd rotational speed omega _r

(5) with described stator magnetic linkage azimuth θ _sWith rotor position angle θ _rSubtract each other the θ that obtains _s-θ _r, with θ _s-θ _rAs the angle of transformation of Park conversion, to the rotor current i under the described rotor two-phase rest frame _{R α}, i _{R β}Carry out the Park conversion, obtain the rotor current i under the synchronous rotating frame _Rd, i _Rq

(6) with the stator magnetic linkage ψ under the described stator rest frame _s, synchronous angular velocity of rotation ω _s, rotor speed ω _rAnd the rotor current i under the synchronous rotating frame _Rd, i _RqInput intersection amount computing module calculates calculating coupling terms u by this module _Rdc, u _Rqc, and meritorious idle reference quantity

(7) with described meritorious idle reference quantity

With the rotor current i under the described synchronous rotating frame _Rd, i _RqSubtract each other respectively, and after carrying out the PI conversion; Respectively with described coupling terms u _Rdc, u _RqcAddition obtains the rotor voltage reference value under the synchronous rotating frame

With this voltage reference value

With θ _s-θ _rFor angle of transformation carries out anti-Park conversion, obtain the rotor voltage reference value under the rotor two-phase rest frame

(8) with described rotor voltage reference value

The input PWM generator, and according to described DC bus-bar voltage u _DcProduce the PWM control wave, drive the PWM current transformer.

2, non-position sensor vector control method as claimed in claim 1 is characterized in that: described step (4) comprises the steps:

(4A) with the rotor current i under the described rotor two-phase rest frame _{R α}, i _{R β}After the PI conversion, obtain the rotor voltage u under the two-phase rest frame _{R α}, u _{R β}

(4B) with described rotor voltage u _{R α}, u _{R β}Input voltage model flux observation module obtains the rotor flux under the stationary rotor coordinate system

(4C) with the rotor current i under the described rotor two-phase rest frame _{R α}, i _{R β}After polar coordinate transform, obtain the rotor current vector under the stationary rotor coordinate system through rectangular coordinate

(4D) with described rotor current vector With described rotor flux

Pass through calculating formula ${\stackrel{\&RightArrow;}{\mathrm{\&psi;}}}_s^r=\frac{L_s}{L_m}{\stackrel{\&RightArrow;}{\mathrm{\&psi;}}}_s^r-\frac{L_s{L}_r-L_m^2}{L_m}{\stackrel{\&RightArrow;}{i}}_r^r$ Calculate the stator magnetic linkage under the stationary rotor coordinate system

Transform to the stator rest frame by rotor-position then, obtain stator magnetic linkage vector under the adjustable stator rest frame

(4E) with the stator magnetic linkage vector under the described stator rest frame With stator magnetic linkage vector under the described adjustable stator rest frame

Cross product;

(4F) the cross product result that step (4E) is calculated carries out obtaining adjustable rotating speed after the PI conversion

Then to this adjustable rotating speed

Obtain adjustable rotor position angle behind the integration

(4G) should adjustable rotor position angle

As the angle of transformation of adjustable model, repeating step (4C) to (4F) stator magnetic linkage vector under the stator rest frame of adjustable model output

Stator magnetic linkage vector under the stator rest frame of reference model output in the tracking

Till, export above-mentioned adjustable rotating speed

With adjustable rotor position angle

Described adjustable rotating speed

Be rotor speed ω _r, adjustable rotor position angle Be rotor position angle θ _r

3, non-position sensor vector control method as claimed in claim 1 is characterized in that: the voltage model flux observation module in the described step (3) comprises the steps:

(3A) the stator voltage u under the described stator two-phase rest frame _{S α}, u _{S β}Carry out high-pass filtering respectively, obtain the ψ after the high-pass filtering _{S α h}, ψ _{S β h}

(3B) ψ that the last cycle is calculated _{S α}, ψ _{S β}Carry out rectangular coordinate to polar coordinate transform, obtain the stator magnetic linkage vector magnitude ψ under the polar coordinate system _ScAnd angular position theta _Sc

(3C) to stator magnetic linkage vector magnitude ψ _ScAfter carrying out amplitude limiting processing, with the angular position theta that obtains in the step (3B) _ScCarry out the coordinate transform of polar coordinates as the position angle of new stator magnetic linkage vector, obtain the stator magnetic linkage feedback quantity ψ under the stator two-phase rest frame to rectangular coordinate _{S α c}, ψ _{S β c}

(3D) the stator magnetic linkage feedback quantity ψ that obtains of step (3C) _{S α c}, ψ _{S β c}After carrying out low-pass filtering respectively, obtain the ψ after the low-pass filtering _{S α l}, ψ _{S β l}:

(3E) ψ that obtains of step (3A) _{S α h}, ψ _{S β h}The ψ that obtains with step (3D) _{S α l}, ψ _{S β l}After the addition, obtain the stator magnetic linkage ψ in this cycle respectively _{S α}, ψ _{S β}

(3F) the magnetic linkage ψ that obtains by step (3E) _{S α}, ψ _{S β}Formula as calculated ${\mathrm{\&psi;}}_s=\sqrt{{\mathrm{\&psi;}}_{\mathrm{s\&alpha;}}^2+{\mathrm{\&psi;}}_{\mathrm{s\&beta;}}^2}$ Calculate the magnetic linkage amplitude ψ under the rest frame _sMagnetic linkage ψ _{S α}, ψ _{S β}Formula as calculated ${\mathrm{\&theta;}}_s=\arctan \frac{{\mathrm{\&psi;}}_{\mathrm{s\&beta;}}}{{\mathrm{\&psi;}}_{\mathrm{s\&alpha;}}}$ Calculate the flux linkage vector angular position theta _sWith the magnetic linkage amplitude ψ that is calculated _sWith the flux linkage vector angular position theta _sFormula as calculated ${\stackrel{\&RightArrow;}{\mathrm{\&psi;}}}_s={\mathrm{\&psi;}}_s{e}^{{\mathrm{j\&theta;}}_s}$ Calculate the flux linkage vector under the rest frame

4, non-position sensor vector control method as claimed in claim 2 is characterized in that: the voltage model flux observation module in described step (3) and the step (4B) comprises the steps:

(3A) the stator voltage u under the described stator two-phase rest frame _α, u _βCarry out high-pass filtering respectively, obtain the ψ after the high-pass filtering _{α h}, ψ _{β h}

(3B) ψ that the last cycle is calculated _α, ψ _βCarry out rectangular coordinate to polar coordinate transform, obtain the stator magnetic linkage vector magnitude ψ under the polar coordinate system _cAnd angular position theta _c

(3C) to stator magnetic linkage vector magnitude ψ _cAfter carrying out amplitude limiting processing, with the angular position theta that obtains in the step (3B) _cCarry out the coordinate transform of polar coordinates as the position angle of new stator magnetic linkage vector, obtain the stator magnetic linkage feedback quantity ψ under the stator two-phase rest frame to rectangular coordinate _{α c}, ψ _{β c}

(3D) the stator magnetic linkage feedback quantity ψ that obtains of step (3C) _{α c}, ψ _{β c}After carrying out low-pass filtering respectively, obtain the ψ after the low-pass filtering _{α l}, ψ _{β l}:

(3E) ψ that obtains of step (3A) _{α h}, ψ _{β h}The ψ that obtains with step (3D) _{α l}, ψ _{β l}After the addition, obtain the stator magnetic linkage ψ in this cycle respectively _α, ψ _β

(3F) the magnetic linkage ψ that obtains by step (3E) _α, ψ _βFormula as calculated $\mathrm{\&psi;}=\sqrt{{\mathrm{\&psi;}}_{\mathrm{\&alpha;}}^2+{\mathrm{\&psi;}}_{\mathrm{\&beta;}}^2}$ Calculate the magnetic linkage amplitude ψ under the rest frame; Magnetic linkage ψ _α, ψ _βFormula as calculated $\mathrm{\&theta;}=\arctan \frac{{\mathrm{\&psi;}}_{\mathrm{\&beta;}}}{{\mathrm{\&psi;}}_{\mathrm{\&alpha;}}}$ Calculate the flux linkage vector angular position theta; With the magnetic linkage amplitude ψ that calculated and flux linkage vector angular position theta formula as calculated $\stackrel{\&RightArrow;}{\mathrm{\&psi;}}=\mathrm{\&psi;}{e}^{\mathrm{j\&theta;}}$ Calculate the flux linkage vector under the rest frame

Images