CN102291079A - Speed sensor-free control algorithm for direct-drive permanent magnet synchronous wind power generation system - Google Patents

Abstract

The invention discloses a control algorithm for estimating real-time stator resistance and speed values of a wind generator by detecting current and voltage and controlling the permanent magnet synchronous wind driven generator by taking the speed estimated by the algorithm as speed feedback. The speed value is estimated by the following steps of: online recognizing the stator resistance value of the wind driven generator based on a model reference adaptation system (MRAS) according to detected stator current and voltage values of the generator, obtaining a computational formula for an error between an estimated electric angle and an actual electric angle of a rotor based on a counter electromotive force and estimating the rotating speed of a fan by eliminating the error. The current and voltage values in the computational formula for the electric angle error are obtained according to the detection, the variable stator resistance value is recognized according to the MRAS in the running of the generator, a differential term of the stator current is realized by adopting a tracking differentiator, thereby more accurately and rapidly estimating the rotating speed.

Description

The synchronous wind power system Speedless sensor of direct-drive permanent-magnetism control algolithm

Technical field

The present invention is the Speedless sensor control algolithm of the synchronous wind power system of a kind of direct-drive permanent-magnetism, adopt the method for stator resistance identification to improve Speedless sensor control algolithm velocity estimation precision and vector control performance, belong to the technical field of wind power generation control appliance.

Background technology

In the synchronous wind power system of direct-drive permanent-magnetism, the vector control of magneto alternator (PMSG) needs the speed and the positional information of rotor, the most frequently used method is a sensor installation on armature spindle, but when the installation of transducer has reduced system reliability and increased cost, brought problem also for the maintenance of system.And general sensor for temperature, humidity, electromagnetic environment also all have requirement, and wind energy conversion system is arranged on the abominable field of environment usually, makes that general working sensor is unreliable.And PMSG rotor mechanical angular speed is very low in the more powerful directly driven wind-powered system, and traditional transducer is difficult to accurately locate electrical degree, thereby is difficult to reach fine control performance.But the PMSG number of pole-pairs is more, and motor electric angle speed can be very not low when the rotor low cruise, thereby stator voltage can be very not low yet, and the influence that the electric angle Speed identification was subjected to when mechanical separator speed was low is less; And do not require Speed identification under the zero-speed state in that system is in service.Therefore,, reduce the maintenance cost of system, in directly driven wind-powered system, generally all save encoder uniform velocity transducer, adopt the Speedless sensor control technology in order to put forward the reliability of wind-powered electricity generation unit operation.Permagnetic synchronous motor Speedless sensor control technology mainly can be divided into following 3 classes at present: a class is based on the open loop computational methods of motor ideal model, the open-loop algorithm that is based on motor mathematical model that adopts, estimation algorithm that changes as direct computing method, based on inductance and back electromotive force integration method etc.This class methods computational process is simple, can access rotor position estimation result comparatively accurately under the correct prerequisite of the parameter of electric machine.But because it need measure the current/voltage value exactly, and has very strong dependence to the parameter of electric machine, add that it is a kind of open loop Calculation Method, can't guarantee that motor is being subjected to noise jamming or still can obtains correct result when parameter changes; The another kind of closed loop algorithm that is based on various observer model, Chinese scholars has been studied the closed loop algorithm based on various observers, current application has extended Kalman filter, sliding mode observer, model reference adaptive algorithm and other adaptive algorithm etc. comparatively widely, and the permagnetic synchronous motor Speedless sensor is controlled on the stability of speed accuracy of observation and system and all is greatly improved.All these are applicable to that the no transducer control technology of high speed operation all is directly or indirectly based on extracting position signal from back-emf, and the voltage equation by motor calculates the electromotive force of being responded to and carries out the estimation of rotor-position.Because the back-emf amplitude is directly proportional with electric angle speed, when the electric angle rotating speed very low in addition when the zero-speed signal to noise ratio of back electromotive force little, other disturbing factor in addition, this class algorithm can not accurately be estimated spinner velocity and position, but more powerful direct wind-driven generator number of pole-pairs is more, even electric angle speed is not low yet when low rate start, and operates in certain range of speeds, the influence that the rotating speed identification is subjected to is less, and therefore this method can be applied in the directly driven wind-powered system; Be to serve as the algorithm based on the motor ideal characterisitics of typical case's representative with the high frequency injection method at last, in order to solve the speed observation problem under the low speed, 1993, M. Corley and R. D. Lorenz have proposed the high-frequency signal method for implanting, measure the variation of motor inductance parameters by injecting high-frequency current signal or high-frequency voltage signal, estimate the low speed and the zero-speed rotor-position of permanent magnet synchronous motor, this method utilizes the salient pole nature of salient-pole machine self or non-salient pole machine showed under high-frequency signal salient pole nature to estimate motor rotor position and rotating speed, is that current permagnetic synchronous motor does not have the effective method of mechanical transducer low cruise.From latter stage the 90's of 20th century so far, had increasing people to study the no transducer control technology of low speed and zero-speed permanent magnet synchronous motor both at home and abroad, and constantly had new method to occur.Avoided direct calculating counter electromotive force of motor based on the method for motor non-ideal characteristic, but obtained rotor position information, all obtained result preferably at motor low speed even zero-speed with the excitation of characteristic frequency.But motor needs the excitation that continues in running, reduced the voltage utilization of inverter, and wind power generator control system does not start when low speed and zero-speed.Because this class methods signal processing is comparatively complicated, dynamic property is not very good in addition.And the torque pulsation that high-frequency signal produced can cause damage to the power transmission shaft of PMSG, and is inapplicable in directly driven wind-powered system.

Summary of the invention

Variation causes the coarse problem of rotating speed identification at parameters such as motor stator resistance, the invention provides the synchronous wind power system Speedless sensor of a kind of direct-drive permanent-magnetism control algolithm.It is to contain the equation that remains to estimate the stator resistance parameter as adjustable model that the identification of MRAS stator resistance gets main thought, and as the reference model, two models have the output variable of identical physical significance with the equation that do not contain unknown parameter.Two models are worked simultaneously, and the stator resistance parameter of utilizing the difference of its output variable to come the real-time regulated adjustable model according to adaptive rate, follow the tracks of the purpose of actual resistance to reach the stator resistance identifier.Adaptive rate need satisfy stability of a system principle, and the asymptotic convergence of system and speed is guaranteed by the Popov superstability.The present invention has utilized the PMSG model parameter so that control effect and control cost and reach comparatively ideal coordination, adopts stator resistance parameter identification algorithm based on MRAS to solve parameter such as motor stator resistance and changes the coarse problem of rotating speed identification that causes.The stator resistance parameter that utilization picks out obtains rotor based on back electromotive force and estimates electrical degree and actual electrical angle calculation formula, and the stator current differential term in the formula has then adopted tracking differentiator (TD) to realize.Estimate that by eliminating rotor electrical degree and rotor actual electrical angular error estimate the wind energy conversion system rotating speed, utilize the estimation rotating speed wind-driven generator to be realized closed-loop vector control as feedback.

Technical solution of the present invention can be divided into following step and realize:

(1) current/voltage coordinate transform, the aerogenerator stator current value and magnitude of voltage and the estimated motor electrical degree that obtain according to detection are that the benchmark of rotational coordinates carries out the dq rotating coordinate transformation to stator current and voltage with the axis of rotor permanent magnetic potential.

(2) stator resistance identification utilizes the current/voltage value behind the rotating coordinate transformation, based on MRAS the motor stator resistance parameter of real-time change is picked out.

(3) rotating speed is estimated, be used to receive the motor stator resistance parameter of current value after the coordinate transform and magnitude of voltage, on-line identification, obtain rotor based on back electromotive force and estimate electrical degree and actual electrical angle calculation formula, the current/voltage value obtains according to detection in the computing formula, identification obtains the stator resistance value according to MRAS, and the stator current differential term has then adopted tracking differentiator (TD) to realize.Estimate that by eliminating rotor electrical degree and rotor actual electrical angular error estimate the wind energy conversion system rotating speed, utilize the estimation rotating speed wind-driven generator to be realized closed-loop vector control as feedback.

Advantage of the present invention: if the rotating speed discrimination method does not carry out on-line identification to the parameter of electric machine, then can rotating speed occur because of variations in temperature and the identification of rotor space position is inaccurate, thereby influence the closed-loop drive performance of PMSG.The present invention proposes a kind of based on MRAS permanent-magnetic synchronous motor stator resistance identification scheme, be applied in the synchronous wind power control system of high-power direct-drive permanent-magnetism, for raising the efficiency, reference model selects motor body, adjustable model selects voltage model, the stator resistance value adopts the PI adaptive rate to reach the accurate estimation of stator resistance as the adjustable parameter of adjustable model.After stator resistance on-line identification goes out, calculate two components of back electromotive force under the dq rotational coordinates based on motor mathematical model, calculate estimation electrical degree and actual electrical angular error based on the parameter of electric machine of measuring and estimate, eliminate this error by the PI control law and extract rotor-position and tach signal.The present invention adopts the stator resistance parameter identification algorithm based on MRAS to solve parameter such as motor stator resistance to change the coarse problem of rotating speed identification that causes.Thereby weakened the influence of image parameter variation to a great extent to the rotating speed identification.The stability and the reliability of Speedless sensor control algolithm have been improved.

Description of drawings

Fig. 1 stator resistance identification algorithm block diagram

Fig. 2 Speedless sensor algorithm block diagram

Among the figure 1 is magneto alternator, the 2nd, electric current coordinate converter, the 3rd, voltage coordinate converter, the 4th, stator current adjustable model, the 5th, adaptive control laws, the 6th, back electromotive force estimation, the 7th, error angle estimation, the 8th, PI controller, the 9th, low pass filter, the 10th, integrator.

Symbol and variable declaration among the figure

, ,

The stator three-phase voltage;

,

, The stator three-phase current;

,

Stator voltage on the dq coordinate system;

,

Stator current on the dq coordinate system;

,

The stator dq electric current estimated value of adjustable model;

The stator resistance estimated value of adjustable model;

The estimation rotating speed;

The rotor estimation error;

The estimated rotor position.

Embodiment

Analyze with the non-salient pole permagnetic synchronous motor, the stator current Mathematical Modeling under rotating coordinate system is

(1)

The estimation stator current Mathematical Modeling that constructing variable is adjustable is

(2)

, Be the stator voltage on the dq coordinate system;

,

Be the stator current on the dq coordinate system;

,

Stator dq electric current estimated value for adjustable model;

Be the stator equivalent inductance on the dq coordinate system;

Be stator resistance;

Stator resistance estimated value for adjustable model;

Be the rotor permanent magnet magnetic linkage;

Be differential operator;

Be rotor electric angle speed;

Definition generalized error vector is

，

，

Formula (2) deducts formula (1),

(3)

In the formula

,

Get D= I, then v= De= e

By the overstable theory of Popov, system stability need be satisfied:

(1) linear element Transitive Matrices G( s)= D( SI AM) ^-1Just real for strictness, this proves easily.

(2) the nonlinear time-varying link satisfies the Popov integral inequality, needs to select the suitable parameters adaptive rate to satisfy.

(4)

Be a limited positive constant, do not rely on

Will v, ω substitution integral inequality,

(5)

Press the common configuration of model reference adaptive rate,

Get following proportional integral form

(6)

Formula (6) substitution formula (5) can get

(7)

(8)

(9)

Want

Set up, can select

, ,

,

For with

The limited positive number that same nature is arranged.

For formula (8), get

，?

， >0 (10)

Then

(11)

To formula (10) both sides differentiate:

，

>0 (12)

To formula (9), desirable ,

0, make

Set up.

According to selected adaptive law, can guarantee

Set up, can get

Identification algorithm is

，

，

(13)

Stator resistance identification algorithm block diagram such as Fig. 1.

Because the rotor-position of control system without position sensor is not to obtain by actual detected, and is based on the result of stator resistance identifier

Construct that motor mathematical model estimates, so in the actual rotor position

And estimated position

Between can have error

Suppose that motor rotor position and the physical location estimated differ

, at this moment, the back electromotive force that is produced by rotor permanent magnet has produced in the coordinate with the estimated rotor position orientation

Two components of coordinate system, promptly

With If the directed coordinate of actual rotor position is

Coordinate system, and based on estimated position

Be

Coordinate system.

Can obtain based on the estimation rotor-position by coordinate transform With the estimation rotating speed The permagnetic synchronous motor Mathematical Modeling

(14)

Back electromotive force under the coordinate system is:

(15)

(16)

It is back EMF coefficient.Suppose to estimate that rotating speed and actual speed error are enough little, then back electromotive force is exactly a site error function.Calculate the rotor estimation error

:

(17)

In the rotor estimation error

The differential term that pair electric current is arranged in the computing formula (17) as can be seen.Differential is realized relatively difficulty in practice.Differential term can be ignored in invariablenes turning speed and load variations that does not have and disturbance, but change in reference rotation velocity and to cause current value to change, perhaps when changing, applied load also can cause current value to change, this moment, dq shaft current differential term had to consider, otherwise dynamic property and disturbance rejection ability that rotating speed is estimated will descend.And in wind power control system, change of wind velocity is frequent, thereby the blower fan torque that wind-force forms is all changing constantly, simultaneously blower fan also need according to wind speed adjust dq shaft current value to change wind speed so that the wind energy of catching reaches best, so the speed dynamic of wind-driven generator is estimated very important.To this, adopt and follow the tracks of the differential that differentiator is realized dq shaft current item.

The main thought that d axle or q shaft current are followed the tracks of differentiator (TD) is: with d axle or q shaft current

Be input signal, it will export 2 signals

With

, wherein Follow the tracks of

, and

=

Thereby, As " approximate differential ".Appropriate structuring is followed the tracks of the internal structure of differentiator, signal

,

Can be at input end signal discontinuous or extract continuous signal and differential signal thereof preferably when comprising certain noise, and speed is fast, the precision height.The dq shaft current is followed the tracks of differentiator and is adopted the way of realization of discrete system time-optimal control comprehensive function algorithm as follows:

(18)

(19)

In formula (18), (19)

Be the sampling period,

Be that the speed density factor is followed the tracks of in reflection.The tracking differentiator of this form all has effect well at aspects such as tracking performance, differential quality and elimination electric shocks.Can obtain dq shaft current differential signal with the method.In the high-power wind turbine system, blower fan is in the permanent environment that becomes of a wind speed, current value is not only very big, the current transients value also may be very big, and change very frequent, dq shaft current differential signal molecule denominator proportion in formula (17) arctan function can not be ignored, and accurately follows the tracks of the rotation speed of fan variation fast and and must obtain dq shaft current differential signal, has realized dq shaft current differential function well and follow the tracks of differentiator.

The rotor estimation error

All can obtain after the estimation of each parameter by detection or preceding method in the formula, thereby can estimate error angle

, by right

Carry out obtaining estimating rotating speed after the proportional integral adjusting , to estimating that thereby rotating speed carries out integration and obtains the estimated rotor position

Bigger by the fluctuation of speed that calculates, therefore output adds the one-level low pass filter usually.Speedless sensor algorithm structure such as Fig. 2.

The transfer function that can derive this algorithm by its principle is:

(20)

This is a second-order system, can draw the PI parameter according to response characteristic.

Claims (4)

1. the synchronous wind power system Speedless sensor of direct-drive permanent-magnetism control algolithm may further comprise the steps:

(1) current/voltage coordinate transform, the aerogenerator stator current value and magnitude of voltage and the estimated motor electrical degree that obtain according to detection are that the benchmark of rotational coordinates carries out the dq rotating coordinate transformation to stator current and voltage with the axis of rotor permanent magnetic potential.

(2) stator resistance identification utilizes the current/voltage value behind the rotating coordinate transformation, based on MRAS the motor stator resistance parameter of real-time change is picked out.

(3) rotating speed is estimated, be used to receive the motor stator resistance parameter of current value after the coordinate transform and magnitude of voltage, on-line identification, obtain rotor based on back electromotive force and estimate electrical degree and actual electrical angle calculation formula, the current/voltage value obtains according to detection in the computing formula, identification obtains the stator resistance value according to MRAS, and the stator current differential term has then adopted tracking differentiator (TD) to realize.Estimate that by eliminating rotor electrical degree and rotor actual electrical angular error estimate the wind energy conversion system rotating speed, utilize the estimation rotating speed wind-driven generator to be realized closed-loop vector control as feedback.

2. the synchronous wind power system Speedless sensor of direct-drive permanent-magnetism according to claim 1 control algolithm, it is characterized in that described step (1) is for sending power-on command when wind power system, beginning was from detecting after system received the current transformer power-on command, start the net side converter, temporary transient starter motor side converter not, generator unit stator has only magnitude of voltage, according to the magnitude of voltage that records by phase-locked electrical degree and the electric angle speed that calculates wind-driven generator.According to electrical degree and the electric angle speed calculated, the generator converter controller adopts Speedless sensor closed-loop vector control method starter motor side converter, begin to detect simultaneously the input as coordinate converter of stator voltage and electric current after the startup, the value after the conversion is again as the input of MRAS stator resistance identifier and rotating speed estimator.

3. the synchronous wind power system Speedless sensor of direct-drive permanent-magnetism according to claim 1 control algolithm is characterized in that beginning to start based on MRAS permanent-magnetic synchronous motor stator resistance identification scheme after described step (2) is for the wind-driven generator operation.Reference model selects motor body, and adjustable model selects voltage model, and the stator resistance value adopts the PI adaptive rate to reach the accurate estimation of stator resistance as the adjustable parameter of adjustable model.Analyze with the non-salient pole permagnetic synchronous motor, the stator current Mathematical Modeling under rotating coordinate system is

$p\left[\begin{array}{c}{i}_d\\ i_q\end{array}\right]=\left[\begin{array}{cc}-\frac{R_s}{L_s}& \mathrm{\ω}\\ -\mathrm{\ω}& -\frac{R_s}{L_s}\end{array}\right]\left[\begin{array}{c}{i}_d\\ i_q\end{array}\right]+\left[\begin{array}{c}{u}_d\\ u_q-{\mathrm{\ψ}}_f\·\mathrm{\ω}\end{array}\right]---\left(1\right)$

The estimation stator current Mathematical Modeling that constructing variable is adjustable is

$p\left[\begin{array}{c}{\hat{i}}_d\\ {\hat{i}}_q\end{array}\right]=\left[\begin{array}{cc}-\frac{{\hat{R}}_s}{L_s}& \mathrm{\ω}\\ -\mathrm{\ω}& -\frac{{\hat{R}}_s}{L_s}\end{array}\right]\left[\begin{array}{c}{\hat{i}}_d\\ {\hat{i}}_q\end{array}\right]+\left[\begin{array}{c}{u}_d\\ u_q-{\mathrm{\ψ}}_f\·\mathrm{\ω}\end{array}\right]---\left(2\right)$

u _d, u _qBe the stator voltage on the dq coordinate system;

i _d, i _qBe the stator current on the dq coordinate system;

Stator dq electric current estimated value for adjustable model;

L _sBe the stator equivalent inductance on the dq coordinate system;

R _sBe stator resistance;

Stator resistance estimated value for adjustable model;

ψ _fBe the rotor permanent magnet magnetic linkage;

P is a differential operator;

ω is a rotor electric angle speed.

Definition generalized error vector is

$e=i-\hat{i}=\left[\begin{array}{c}{i}_d-{\hat{i}}_d\\ i_q-{\hat{i}}_q\end{array}\right],$ $i=\left[\begin{array}{c}{i}_d\\ i_q\end{array}\right],$ $\hat{i}=\left[\begin{array}{c}{\hat{i}}_d\\ {\hat{i}}_q\end{array}\right]$

Press the common configuration of model reference adaptive rate, select adaptive law for use according to the overstable theory of Popov,

Identification algorithm is

${\hat{R}}_s=k_i{\∫}_0^t({\hat{i}}_d^2-i_d{\hat{i}}_d+{\hat{i}}_q^2-i_q{\hat{i}}_q)\mathrm{d\τ}+k_p({\hat{i}}_d^2-i_d{\hat{i}}_d+{\hat{i}}_q^2-i_q{\hat{i}}_q)+{\hat{R}}_s\left(0\right)---\left(3\right)$

4. the synchronous wind power system Speedless sensor of the direct-drive permanent-magnetism control algolithm based on the identification of stator resistance according to claim 1, it is characterized in that described step (3) rotating speed estimates to utilize the motor stator resistance parameter of current value after the coordinate transform and magnitude of voltage, on-line identification, estimate that by eliminating rotor electrical degree and rotor actual electrical angular error estimate the wind energy conversion system rotating speed.Suppose the rotor estimated position

θ differs θ with the actual rotor position _e, the back electromotive force that is produced by rotor permanent magnet has produced in the coordinate with the estimated rotor position orientation

Two components of coordinate system With

If the directed coordinate of actual rotor position is d _qCoordinate system, and based on estimated position

Be Coordinate system.Can obtain based on the estimation rotor-position by coordinate transform

With the estimation rotating speed

The permagnetic synchronous motor Mathematical Modeling

$\left\{\begin{array}{c}{\hat{u}}_d={\hat{R}}_s{\hat{i}}_d+L_{s}p{\hat{i}}_d-\widehat{\mathrm{\ω}}{L}_s{\hat{i}}_q+{\hat{e}}_d\\ {\hat{u}}_q={\hat{R}}_s{\hat{i}}_q+L_{s}p{\hat{i}}_q+\widehat{\mathrm{\ω}}{L}_s{\hat{i}}_d+{\hat{e}}_q\end{array}\right.---\left(4\right)$

Back electromotive force under the coordinate system is:

$\left\{\begin{array}{c}{\hat{e}}_d=-E_{\mathrm{ex}}\sin {\mathrm{\θ}}_e-(\widehat{\mathrm{\ω}}-\mathrm{\ω})L_s{i}_q\\ {\hat{e}}_q=E_{\mathrm{ex}}\cos {\mathrm{\θ}}_e+(\widehat{\mathrm{\ω}}-\mathrm{\ω})L_s{i}_d\end{array}\right.---\left(5\right)$

E _ex＝ω·ψ _f (6)

E _ExIt is back EMF coefficient.Suppose to estimate that rotating speed and actual speed error are enough little, then back electromotive force is exactly a site error function.Calculate the rotor estimation error

${\widehat{\mathrm{\θ}}}_e={\tan }^{-1}\left(\frac{-{\hat{e}}_d}{{\hat{e}}_q}\right)$

$={\tan }^{-1}\frac{-({\hat{u}}_d-{\hat{R}}_s{\hat{i}}_d+L_s\widehat{\mathrm{\ω}}{\hat{i}}_q-L_{s}p{\hat{i}}_d)}{{\hat{u}}_q-{\hat{R}}_s{\hat{i}}_q-L_s\widehat{\mathrm{\ω}}{\hat{i}}_d-L_{s}p{\hat{i}}_q}---\left(7\right)$

In the rotor estimation error

Computing formula has the current differential item in (7).In wind power control system, change in reference rotation velocity and to cause current value to change, perhaps also can cause current value to change when applied load changes, this moment, dq shaft current differential term can not be ignored, otherwise dynamic property and disturbance rejection ability that rotating speed is estimated will descend.Adopt and follow the tracks of the differential that differentiator is realized dq shaft current item.The structure that d axle or q shaft current are followed the tracks of differentiator (TD) is: d axle or q shaft current i (t) are input signal, and it will export 2 signal x ₁And x ₂, x wherein ₁Follow the tracks of i (t), and

Thereby x ₂" approximate differential " as i (t).Appropriate structuring is followed the tracks of the internal structure of differentiator, signal x ₁, x ₂Can be at input end signal discontinuous or extract continuous signal and differential signal thereof preferably when comprising certain noise, and speed is fast, the precision height.The dq shaft current is followed the tracks of differentiator and is adopted the way of realization of discrete system time-optimal control comprehensive function algorithm as follows:

$\left\{\begin{array}{c}{x}_1(k+1)=x_1\left(k\right)+{\mathrm{hx}}_2\left(k\right)\\ x_2(k+1)=x_2\left(k\right)+\mathrm{hf},\left|f\right|\≤r\\ f=\mathrm{fhan}(x_1\left(k\right),x_2\left(k\right),r,h)\end{array}\right.---\left(8\right)$

$\left\{\begin{array}{c}d=\mathrm{rh}\\ d_0=\mathrm{hd}\\ y=x_1+h{x}_2\\ d_0=\sqrt{d^2+8r\left|y\right|}\\ a=\left\{\begin{array}{c}{x}_2+\frac{(a_0-d)}{2}\·\mathrm{sign}\left(y\right),\left|y\right|>d_0\\ x_2+\frac{y}{h},\left|y\right|\≤d_0\end{array}\right.\\ \mathrm{fhan}=\left\{\begin{array}{c}r\·\mathrm{sign}\left(a\right),\left|a\right|>d\\ r\frac{a}{d},\left|a\right|\≤d\end{array}\right.\end{array}\right.----\left(9\right)$

H is the sampling period in formula (8), (9), and r is that the speed density factor is followed the tracks of in reflection.Obtain dq shaft current differential signal with the method.

The rotor estimation error

Obtain after the estimation of each parameter by detection or preceding method in the formula, thereby estimate error angle

By right

Carry out obtaining the initial estimation rotating speed after the PI adjusting, finally estimated rotating speed through the one-level low pass filter

To estimating that thereby rotating speed carries out integration and obtains the estimated rotor position

Images