CN103701386A - Flux linkage error observation-based acquisition method of full-order flux linkage observer of asynchronous motor without speed sensor - Google Patents

Abstract

The invention discloses a flux linkage error observation-based acquisition method of a full-order flux linkage observer of an asynchronous motor without a speed sensor and belongs to the field of a speed sensorless vector control full-order flux linkage observer. The problem that the existing speed sensorless vector control system causes low observation accuracy of the full-order flux linkage observer due to larger errors of motor parameters when a motor runs at low speed, and finally, the running stability of the system is poor is solved. A full-order flux linkage observer error feedback matrix coefficient is obtained according to the following rules, namely, the pole real part of the observer is smaller than the pole real part of an asynchronous motor, the real parts are both negative numbers, the zero pole real parts of an estimation rotation speed and a transfer function are both negative numbers, the error between an estimation flux linkage and a real flux linkage is utilized, when the motor runs at low speed, the equivalence of the system is a current model, and when the motor runs at high speed, the equivalence of the system is a voltage model. The rotor flux linkage phase position error coefficient ilambda is utilized and the rotor flux linkage amplitude error coefficient k is introduced, so that the estimation rotating speed precision is increased. The flux linkage error observation-based acquisition method of the full-order flux linkage observer of the asynchronous motor without the speed sensor is particularly used in the field of speed sensorless vector control.

Description

The acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error

Technical field

The invention belongs to flux observer field, the full rank of Speedless sensor vector control.

Background technology

Vector Control System of Induction Motor technology can realize the decoupling zero of torque and magnetic linkage, and has good dynamic characteristic and steady-state characteristic, so obtained application very widely in industrial system.In a lot of industrial occasions, require motor can stable operation in low rotation speed area, as elevator, hoist engine, excavator etc., but because the required speed probe of control is expensive and very easily damage, so reduced the reliability of governing system, increased maintenance cost.And speed-less sensor vector control system is when low cruise, because parameter of electric machine error is larger, be easy to cause system fluctuation of service.To sum up, for fear of using speed probe, in the useful life that strengthens system, be necessary to carry out the research of Speedless sensor vector control low-speed stable operating scheme.

Speedless sensor vector control, according to observer principle, is utilized the equations of state that asynchronous machine dynamic model forms to estimate stator and rotor flux, and is introduced the accuracy of observation that Error Feedback improves state variable.Owing to comprising rotor speed variable information in observer, therefore can be according to observer principle design rotating speed adaptive law observation rotor speed.But flux observation accuracy and speed observation accuracy and the parameter of electric machine are closely related, when the parameter of electric machine is inaccurate, can move motor stabilizing, especially motor causes very large impact when low cruise.When motor operates in high speed, counter electromotive force of motor is very large, so parameter is relatively little on the impact of control system, speed-less sensor vector control system can keep stable operation.But when motor operates in low speed (below 30rpm), counter electromotive force of motor is less, it is large that the impact of the parameter of electric machine becomes, if parameter is inaccurate, can cause magnetic linkage and rotor speed to be estimated inaccurate, causes to control and lost efficacy.Non-synchronous motor parameter can not accurately obtain in real work, and after motor long-play, also can there is larger variation in the parameter of electric machine, so control, motor can be realized good rotary speed precision when low speed and utmost point low speed (below 15rpm) operation and stability tool acquires a certain degree of difficulty.

From current existing speed sensorless vector control technology, the observation procedure of magnetic linkage is mainly divided into following two kinds: 1) open loop flux observation.Open loop flux observation is to take motor dynamics equation as basic flux linkage calculation method, can be divided into voltage model method, the full rank of current model method and open loop flux observer, voltage model method is owing to comprising stator resistance parameters, so when inapplicable and motor low cruise, similarly, current model is inapplicable when the high speed operation of motor.And the full rank of open loop flux observer is owing to there is no Error Feedback compensation term, so system robustness is poor.2) closed loop flux observation.Closed loop flux observation is compared to open loop flux observation system and has introduced Error Feedback item, has improved the robustness of system.Can be divided into model reference adaptive, the full rank of closed loop flux observer, closed loop depression of order flux observer and Kalman filter method.What application was more at present is the full rank of closed loop flux observers.For the method, the problem mainly solving is flux observation accuracy and stability of a system problem, and this need to meet the demands by appropriate design Error Feedback matrix and rotating speed adaptive law.But in prior art, only carry out single satisfied observation magnetic linkage accuracy requirement by design error feedback matrix, or stability of a system requirement.The method for designing that can simultaneously meet two kinds of requirements has no report.

Summary of the invention

The present invention is in order to solve existing speed-less sensor vector control system when the motor low cruise, because parameter of electric machine error is larger, cause the observation accuracy of full rank flux observer low, finally cause the poor problem of system run all right, the invention provides a kind of acquisition methods of full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error.

The acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error, the method realizes based on existing full rank flux observer, it is characterized in that, and the method comprises the steps,

Step 1, below meeting, during 3 conditions, obtain 4 error feedback coefficients, and these 4 error feedback coefficients are respectively g ₁, g ₂, g ₃and g ₄, by 4 error feedback coefficient substitutions of obtaining

$G={\left[\begin{array}{cccc}{g}_1& {-g}_2& g_3& {-g}_4\\ g_2& g_1& g_4& g_3\end{array}\right]}^T$ (formula 1)

In, obtain G,

Wherein, G represents the Error Feedback matrix of observer,

3 conditions are respectively,

Condition one: observer limit real part is less than asynchronous machine limit real part, and be all negative,

Condition two: the zero limit real part of estimating rotating speed transfer function is all negative,

Condition three: utilize the error of estimating magnetic linkage and true magnetic linkage, assurance system, when motor low cruise, is equivalent to current model, and system, when high speed operation of motor, is equivalent to voltage model;

Step 2, according to known speed adaptive law equation:

${\widehat{\mathrm{\ω}}}_r=k_1(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})-k_2(e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})$ (formula 2),

Obtain the rear rotating speed adaptive law equation of distortion

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3),

Wherein,

represent the motor speed of estimating,

K ₁represent stator current Error Gain,

E _{i α}expression estimation stator current is compared the error component of transverse axis under rest frame with actual stator electric current,

represent the longitudinal axis component of estimated rotor magnetic linkage under rest frame,

E _{i β}expression estimation stator current is compared the error component of the longitudinal axis under rest frame with actual stator electric current,

the quadrature component of statement estimated rotor magnetic linkage under rest frame,

K ₂represent rotor flux Error Gain,

E _{λ α}expression estimated rotor magnetic linkage is compared the error component of transverse axis under rest frame with actual rotor magnetic linkage,

represent the longitudinal axis component of estimated rotor magnetic linkage under rest frame,

E _{λ β}expression estimated rotor magnetic linkage is compared the error component of the longitudinal axis under rest frame with actual rotor magnetic linkage,

represent the quadrature component of estimated rotor magnetic linkage under rest frame,

K _prepresent the proportional gain of pi controller,

K _ithe storage gain that represents pi controller,

I _sqrepresent the longitudinal axis component of actual stator electric current under rotating coordinate system,

represent the longitudinal axis component of estimated rotor electric current under rotating coordinate system,

represent the quadrature component of estimated rotor magnetic linkage under rotating coordinate system,

K represents rotor flux amplitude error coefficient,

I _sdrepresent the quadrature component of actual stator electric current under rotating coordinate system,

represent to estimate the quadrature component of stator current under rotating coordinate system,

I _λrepresent rotor flux phase error coefficient,

Step 3, the Error Feedback matrix G of the observer obtaining in step 1 is replaced to the Error Feedback matrix in the flux observer of existing full rank, by rotating speed adaptive law equation after the distortion of obtaining in step 2

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3)

Replace the rotating speed adaptive law in the flux observer of existing full rank, successfully obtain the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error.

In described step 2, according to known speed adaptive law equation:

${\widehat{\mathrm{\ω}}}_r=k_1(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})-k_2(e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})$ (formula 2),

Obtain the rear rotating speed adaptive law equation of distortion

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3)

Detailed process be,

In (formula 2),

$e_{\mathrm{i\α}}=i_{\mathrm{s\α}}-{\hat{i}}_{\mathrm{s\α}}$ (formula 4),

$e_{\mathrm{i\β}}=i_{\mathrm{s\β}}-{\hat{i}}_{\mathrm{s\β}}$ (formula 5),

$e_{\mathrm{\λ\α}}={\mathrm{\λ}}_{\mathrm{r\α}}-{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}$ (formula 6),

$e_{\mathrm{\λ\β}}={\mathrm{\λ}}_{\mathrm{r\β}}-{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}$ (formula 7),

Wherein, i _{s α}represent the transverse axis stator current component actual value under static coordinate,

the estimated value that represents the transverse axis stator current component under static coordinate,

I _{s β}represent the longitudinal axis stator current component actual value under static coordinate, the estimated value that represents the longitudinal axis stator current component under static coordinate,

λ _{r α}represent the transverse axis rotor flux component actual value under static coordinate,

represent the transverse axis rotor flux component estimated value under static coordinate,

λ _{r β}represent the longitudinal axis rotor flux component actual value under static coordinate,

represent the longitudinal axis rotor flux component estimated value under static coordinate,

By (formula 4) to (formula 7) substitution (formula 2)

in, carry out obtaining after abbreviation: $e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}={\mathrm{\λ}}_{\mathrm{r\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-{\mathrm{\λ}}_{\mathrm{r\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=\sqrt{{\mathrm{\λ}}_{\mathrm{r\α}}^2+{\mathrm{\λ}}_{\mathrm{r\β}}^2}\·\sqrt{{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}^2+{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}^2}\left(\frac{{\mathrm{\λ}}_{\mathrm{r\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-{\mathrm{\λ}}_{\mathrm{r\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}}{\sqrt{{\mathrm{\λ}}_{\mathrm{r\α}}^2+{\mathrm{\λ}}_{\mathrm{r\β}}^2}\·\sqrt{{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}^2+{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}^2}}\right)$ (formula 8),

Suppose under rest frame actual rotor flux linkage vector

with estimated rotor flux linkage vector

with the angle of α reference axis be respectively θ and

and θ and

difference be △ θ, therefore, according to rotor flux Vector Rotation speed, equal stator current vector rotary speed, (formula 8) through distortion after obtain:

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=\left|{\stackrel{\→}{\mathrm{\λ}}}_r\right|\·\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|(\sin \widehat{\mathrm{\θ}}\cos \mathrm{\θ}-\cos \widehat{\mathrm{\θ}}\sin \mathrm{\θ})=-\sin \mathrm{\Δ\θ}({\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|}^2+\mathrm{\Δ\λ}|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\left|\right)$ (formula 9)

Wherein,

for the rotor flux amplitude of estimating, for actual rotor flux amplitude, △ λ is the amplitude error of actual rotor magnetic linkage and estimated rotor magnetic linkage, and △ λ is 0,

Order

$k={\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|}^2+\mathrm{\Δ\λ}\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|$ (formula 10),

Wherein, k is rotor flux amplitude error coefficient,

In asynchronous machine, the Space Rotating speed of rotor flux vector, stator magnetic linkage vector stator current vector is consistent, in observer, this three's Space Rotating speed is also consistent, therefore, make the error of the rotor flux anglec of rotation of the actual rotor magnetic linkage anglec of rotation and estimation, equal the error of the stator current anglec of rotation of the actual stator current phasor anglec of rotation and observation, utilize the cosine law to obtain

$\sin \mathrm{\Δ\θ}=\frac{|i_{\mathrm{sq}}{\hat{i}}_{\mathrm{sd}}-i_{\mathrm{sd}}{\hat{i}}_{\mathrm{sq}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 11),

Wherein, represent actual stator current phasor,

represent to estimate stator current vector amplitude,

(formula 10) and (formula 11) is updated in (formula 9), obtains

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=-k\frac{|i_{\mathrm{sq}}{\hat{i}}_{\mathrm{sd}}-i_{\mathrm{sd}}{\hat{i}}_{\mathrm{sq}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 12),

Will

be updated in (formula 12),

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=-k\frac{\left|i_{\mathrm{sq}}\right||i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 13),

Utilize pi regulator (k _p+ k _i∫ dt) replace the k in (formula 2) ₁and k ₂, and (formula 13) is updated in (formula 2), obtain ${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}].$

Described existing full rank flux observer comprises Α, B, C, G, 1/s, rotating speed adaptive rate, angle calculation module, an adder and two subtracters, described Α represents full rank flux observation matrix, B represents voltage input matrix, C represents electric current output matrix, 1/s represents integral operation

Adder, for the error compensating signal of the observation signal of B output voltage signal, Α output and G output is sued for peace, obtains rotor flux differential signal,

1/s, for the rotor flux differential signal of adder output is carried out to integral operation, obtains rotor flux signal, and rotor flux signal is sent to respectively to C, Α, angle calculation module, rotating speed adaptive rate,

C is for output estimation stator current quadrature component under rotating coordinate system

with estimation stator current longitudinal axis component under rotating coordinate system

Wherein, a subtracter is used for quadrature component i under the actual stator electric current rotating coordinate system of input _sdwith quadrature component under estimation stator current rotating coordinate system

differ from, the error signal of the stator current of acquisition quadrature component under rotating coordinate system, and the error signal of this stator current quadrature component under rotating coordinate system is sent to rotating speed adaptive rate and G,

Another subtracter is used for longitudinal axis component i under the actual stator electric current rotating coordinate system of input _sqwith longitudinal axis component under estimation stator current rotating coordinate system

differ from, the error signal of the stator current of acquisition longitudinal axis component under rotating coordinate system, and the error signal of this stator current longitudinal axis component under rotating coordinate system is sent to rotating speed adaptive rate and G simultaneously,

Angle calculation module is used and rotor flux signal is carried out to angle calculation, and Α is used for exporting observation signal, and rotating speed adaptive rate is used for output speed feedback signal, and this speed feedback signal is sent to Α.

The full rank flux observation implement body of the Speed Sensorless Induction Motor based on observation magnetic linkage error that the present invention obtains is applied in universal frequency converter speed-less sensor vector control system, and the logic mechanism schematic diagram of this universal frequency converter speed-less sensor vector control system is specifically referring to Fig. 3.

Method of the present invention is that the rotor flux linkage orientation based under two-phase rotating coordinate system carries out, rotation dq coordinate system is consistent with rotor flux Vector Rotation speed, make d axle overlap with rotor flux vector, and calculate the anglec of rotation of rotor flux vector, utilize this angle to carry out the conversion between stator current three phase static coordinate system and two-phase rotating coordinate system, transformation for mula is as follows:

$\left[\begin{array}{c}{i}_{\mathrm{s\α}}\\ i_{\mathrm{s\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{i}_U\\ i_V\\ i_W\end{array}\right]$ With $\left[\begin{array}{c}{i}_{\mathrm{sd}}\\ i_{\mathrm{sq}}\end{array}\right]=\left[\begin{array}{cc}\cos \mathrm{\θ}& -\sin \mathrm{\θ}\\ \sin \mathrm{\θ}& \cos \mathrm{\θ}\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{s\α}}\\ i_{\mathrm{s\β}}\end{array}\right]$

Wherein, θ is the rotor field anglec of rotation of utilizing estimated rotor flux linkage calculation to go out, and its computing formula is:

Wherein, i _{s α}represent quadrature component under actual stator electric current rest frame, i _{s β}represent longitudinal axis component under actual stator electric current rest frame, i _urepresent threephase stator electric current first-phase, i _vrepresent threephase stator electric current second-phase, i _wrepresent threephase stator electric current third phase,

The dq shaft current of finally utilizing current-order and sampling to decompose forms the current inner loop of vector control system, and output order voltage, after SVPWM modulation, produces turning on and off of switching signal control switch pipe, finally reaches the object of frequency control.

Utilize universal frequency converter speed-less sensor vector control system to drive and control induction machine, the parameter of this induction machine is as follows: rated voltage: 380V, rated current 15.4A, rated power is 7.5Kw, rated speed is 1440r/min, and rated frequency is 50Hz, and it is 5000P/R that rotating speed detects encoder line number, after 1028 segmentations, number of buses is 5 * 10 ⁶;

Rotor speed oscillogram when Fig. 4 is speed sensor vector control, the velocity transducer line number adopting in Fig. 4 is 1000P/R, as can be seen from Figure 4, because load exists the fluctuation of 6% nominal torque and the application of high-precision encoder checkout equipment, even so when speed sensor is controlled, still there is the fluctuation of 3rpm in rotating speed, therefore can reach a conclusion, the motor low-speed performance of speed sensorless vector control is similar to the motor low-speed performance of speed sensor vector control, the validity of the method that the present invention that hence one can see that proposes.Speed waveform figure when Fig. 5 is Speedless sensor vector control, in Fig. 4, speed sensor vector control is compared with Speedless sensor vector control in Fig. 5, and the fluctuation of speed does not have that great changes will take place,

Fig. 6 is under Speedless sensor vector control condition, when nominal torque that the fluctuation of load is 34%, rotor speed oscillogram, utilize when nominal torque that the fluctuation of load is 34%, the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error that the present invention is obtained, carry out disturbance rejection experiment, system still can keep stable at utmost point low cruise state, specifically referring to Fig. 7 to Figure 10, in Fig. 7 during speed sensor vector control, 1.5rpm rotary speed instruction 100% nominal load, the fluctuation of load is 6%, in Fig. 8 during speed sensor vector control, 1.5rpm rotary speed instruction 100% nominal load, the fluctuation of load is 6%, Speedless sensor vector control in Fig. 9, 1.5rpm rotary speed instruction 100% nominal load, the fluctuation of load is 6%, Speedless sensor vector control in Figure 10, 1.5rpm rotary speed instruction 100% nominal load, the fluctuation of load is 6%, Speedless sensor vector control is compared with speed sensor vector control, effect when motor utmost point low cruise is similar, can keeping system stable operation, there is not severe variation.

The present invention is in order to guarantee observation accuracy and the stability of a system of observer simultaneously, according to following criterion, design full rank flux observer Error Feedback matrix coefficient: (1) observer limit real part is less than asynchronous machine limit real part, and be all negative, enable to guarantee that the convergence rate of observation magnetic linkage is greater than the convergence rate of true magnetic linkage; (2) the transfer function zero limit real part of estimation rotating speed is all negative, and assurance estimation rotating speed can be restrained under any system gain; (3) utilize the error of estimating magnetic linkage and true magnetic linkage, assurance system, when motor low cruise, is equivalent to current model, and system, when high speed operation of motor, is equivalent to voltage model, dwindles observation magnetic linkage amplitude error and phase error; And according to the rotor flux phase error coefficient i in the phase angle error design speed adaptive law of estimation magnetic linkage and actual magnetic linkage _λ, introducing rotor flux amplitude error coefficient k increases estimation rotary speed precision.

The beneficial effect that the present invention brings is, owing to utilizing respectively the problem of rotating speed adaptive law and full rank magnetic linkage Error Feedback matrix resolution system magnetic linkage accuracy of estimation and stability, so can guarantee that motor is under the condition of Speedless sensor, (1.5rpm) long-time steady operation under low-down rotating speed.And because the parameter of introducing is less, make the present invention there is stronger versatility.When motor operates in utmost point low speed, the accuracy of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error of the present invention has improved more than 30%, makes the governing system can stable operation.

Accompanying drawing explanation

Fig. 1 is the logical construction schematic diagram of the existing full rank flux observer described in embodiment three;

Fig. 2 is the logical construction schematic diagram of the rear rotating speed adaptive law equation of distortion in embodiment one; Wherein, PI is pi controller, "

" expression subtracter, "

" expression adder,

Fig. 3 is the logical construction schematic diagram of universal frequency converter speed-less sensor vector control system in summary of the invention; Wherein,

for rotary speed instruction signal,

for magnetic linkage command signal,

for quadrature component under stator current instruction rest frame,

for longitudinal axis component under stator current instruction rest frame,

for stator voltage vector,

represent stator voltage vector instruction, SVPWM is space vector pulse width modulation, dq α β represent rotating coordinate transformation, α β abc represent static coordinate conversion;

Rotor speed oscillogram when Fig. 4 is speed sensor vector control;

Rotor speed oscillogram when Fig. 5 is Speedless sensor vector control;

Fig. 6 is under Speedless sensor vector control condition, when nominal torque that the fluctuation of load is 34%, and rotor speed oscillogram;

When Fig. 7 is speed sensor vector control, rotor speed oscillogram when induction machine operates in 1.5rpm;

When Fig. 8 is speed sensor vector control, the oscillogram of stator current when induction machine operates in 1.5rpm, torque current and magnetic linkage electric current;

When Fig. 9 is Speedless sensor vector control, rotor speed oscillogram when induction machine operates in 1.5rpm;

When Figure 10 is Speedless sensor vector control, stator current when induction machine operates in 1.5rpm, torque current and magnetic linkage current waveform figure.

Embodiment

Embodiment one: present embodiment is described referring to Fig. 2, the acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error described in present embodiment, the method realizes based on existing full rank flux observer, and the method comprises the steps

Step 1, below meeting, during 3 conditions, obtain 4 error feedback coefficients, and these 4 error feedback coefficients are respectively g ₁, g ₂, g ₃and g ₄, by 4 error feedback coefficient substitutions of obtaining

$G={\left[\begin{array}{cccc}{g}_1& {-g}_2& g_3& {-g}_4\\ g_2& g_1& g_4& g_3\end{array}\right]}^T$ (formula 1)

In, obtain G,

Wherein, G represents the Error Feedback matrix of observer,

3 conditions are respectively,

Condition one: observer limit real part is less than asynchronous machine limit real part, and be all negative,

Condition two: the zero limit real part of estimating rotating speed transfer function is all negative,

Condition three: utilize the error of estimating magnetic linkage and true magnetic linkage, assurance system, when motor low cruise, is equivalent to current model, and system, when high speed operation of motor, is equivalent to voltage model;

Step 2, according to known speed adaptive law equation:

${\widehat{\mathrm{\ω}}}_r=k_1(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})-k_2(e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})$ (formula 2),

Obtain the rear rotating speed adaptive law equation of distortion

${\widehat{\mathrm{\ω}}}_r=k_1(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})-k_2(e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})$ (formula 3),

Wherein,

represent the motor speed of estimating,

K ₁represent stator current Error Gain,

E _{i α}expression estimation stator current is compared the error component of transverse axis under rest frame with actual stator electric current,

represent the longitudinal axis component of estimated rotor magnetic linkage under rest frame,

E _{i β}expression estimation stator current is compared the error component of the longitudinal axis under rest frame with actual stator electric current,

the quadrature component of statement estimated rotor magnetic linkage under rest frame,

K ₂represent rotor flux Error Gain,

E _{λ α}expression estimated rotor magnetic linkage is compared the error component of transverse axis under rest frame with actual rotor magnetic linkage,

represent the longitudinal axis component of estimated rotor magnetic linkage under rest frame,

E _{λ β}expression estimated rotor magnetic linkage is compared the error component of the longitudinal axis under rest frame with actual rotor magnetic linkage,

represent the quadrature component of estimated rotor magnetic linkage under rest frame,

K _prepresent the proportional gain of pi controller,

K _ithe storage gain that represents pi controller,

I _sqrepresent the longitudinal axis component of actual stator electric current under rotating coordinate system,

represent the longitudinal axis component of estimated rotor electric current under rotating coordinate system,

represent the quadrature component of estimated rotor magnetic linkage under rotating coordinate system,

K represents rotor flux amplitude error coefficient,

I _sdrepresent the quadrature component of actual stator electric current under rotating coordinate system,

represent to estimate the quadrature component of stator current under rotating coordinate system,

I _λrepresent rotor flux phase error coefficient,

Step 3, the Error Feedback matrix G of the observer obtaining in step 1 is replaced to the Error Feedback matrix in the flux observer of existing full rank, by rotating speed adaptive law equation after the distortion of obtaining in step 2

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3)

Replace the rotating speed adaptive law in the flux observer of existing full rank, successfully obtain the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error.

Embodiment two: present embodiment is described referring to Fig. 1, the difference of the acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error described in present embodiment and embodiment one is, in described step 2, according to known speed adaptive law equation:

${\widehat{\mathrm{\ω}}}_r=k_1(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})-k_2(e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})$ (formula 2),

Obtain the rear rotating speed adaptive law equation of distortion

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3)

Detailed process be,

In (formula 2),

$e_{\mathrm{i\α}}=i_{\mathrm{s\α}}-{\hat{i}}_{\mathrm{s\α}}$ (formula 4),

$e_{\mathrm{i\β}}=i_{\mathrm{s\β}}-{\hat{i}}_{\mathrm{s\β}}$ (formula 5),

$e_{\mathrm{\λ\α}}={\mathrm{\λ}}_{\mathrm{r\α}}-{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}$ (formula 6),

$e_{\mathrm{\λ\β}}={\mathrm{\λ}}_{\mathrm{r\β}}-{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}$ (formula 7),

Wherein, i _{s α}represent the transverse axis stator current component actual value under static coordinate,

the estimated value that represents the transverse axis stator current component under static coordinate,

I _{s β}represent the longitudinal axis stator current component actual value under static coordinate,

the estimated value that represents the longitudinal axis stator current component under static coordinate,

λ _{r α}represent the transverse axis rotor flux component actual value under static coordinate,

represent the transverse axis rotor flux component estimated value under static coordinate,

λ _{r β}represent the longitudinal axis rotor flux component actual value under static coordinate,

represent the longitudinal axis rotor flux component estimated value under static coordinate,

By (formula 4) to (formula 7) substitution (formula 2) in, carry out obtaining after abbreviation: $e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}={\mathrm{\λ}}_{\mathrm{r\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-{\mathrm{\λ}}_{\mathrm{r\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=\sqrt{{\mathrm{\λ}}_{\mathrm{r\α}}^2+{\mathrm{\λ}}_{\mathrm{r\β}}^2}\·\sqrt{{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}^2+{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}^2}\left(\frac{{\mathrm{\λ}}_{\mathrm{r\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-{\mathrm{\λ}}_{\mathrm{r\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}}{\sqrt{{\mathrm{\λ}}_{\mathrm{r\α}}^2+{\mathrm{\λ}}_{\mathrm{r\β}}^2}\·\sqrt{{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}^2+{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}^2}}\right)$ (formula 8),

Suppose under rest frame actual rotor flux linkage vector

with estimated rotor flux linkage vector

with the angle of α reference axis be respectively θ and

and θ and

difference be △ θ, therefore, according to rotor flux Vector Rotation speed, equal stator current vector rotary speed, (formula 8) through distortion after obtain: $e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=\left|{\stackrel{\→}{\mathrm{\λ}}}_r\right|\·\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|(\sin \widehat{\mathrm{\θ}}\cos \mathrm{\θ}-\cos \widehat{\mathrm{\θ}}\sin \mathrm{\θ})=-\sin \mathrm{\Δ\θ}({\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|}^2+\mathrm{\Δ\λ}|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\left|\right)$ (formula 9)

Wherein,

for the rotor flux amplitude of estimating,

for actual rotor flux amplitude, △ λ is the amplitude error of actual rotor magnetic linkage and estimated rotor magnetic linkage, and △ λ is 0,

Order

$k={\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|}^2+\mathrm{\Δ\λ}\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|$ (formula 10),

Wherein, k is rotor flux amplitude error coefficient,

In asynchronous machine, the Space Rotating speed of rotor flux vector, stator magnetic linkage vector stator current vector is consistent, in observer, this three's Space Rotating speed is also consistent, therefore, make the error of the rotor flux anglec of rotation of the actual rotor magnetic linkage anglec of rotation and estimation, equal the error of the stator current anglec of rotation of the actual stator current phasor anglec of rotation and observation, utilize the cosine law to obtain

$\sin \mathrm{\Δ\θ}=\frac{|i_{\mathrm{sq}}{\hat{i}}_{\mathrm{sd}}-i_{\mathrm{sd}}{\hat{i}}_{\mathrm{sq}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 11),

Wherein,

represent actual stator current phasor,

represent to estimate stator current vector amplitude,

(formula 10) and (formula 11) is updated in (formula 9), obtains

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=-k\frac{|i_{\mathrm{sq}}{\hat{i}}_{\mathrm{sd}}-i_{\mathrm{sd}}{\hat{i}}_{\mathrm{sq}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 12),

Will

be updated in (formula 12),

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=-k\frac{\left|i_{\mathrm{sq}}\right||i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 13),

Utilize pi regulator (k _p+ k _i∫ dt) replace the k in (formula 2) ₁and k ₂, and (formula 13) is updated in (formula 2), obtain ${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}].$

Embodiment three: present embodiment is described referring to Fig. 3, the difference of the acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error described in present embodiment and embodiment one is, described existing full rank flux observer comprises Α, B, C, G, 1/s, rotating speed adaptive rate, angle calculation module, an adder and two subtracters, described Α represents full rank flux observation matrix, B represents voltage input matrix, C represents electric current output matrix, 1/s represents integral operation

Adder, for the error compensating signal of the observation signal of B output voltage signal, Α output and G output is sued for peace, obtains rotor flux differential signal,

1/s, for the rotor flux differential signal of adder output is carried out to integral operation, obtains rotor flux signal, and rotor flux signal is sent to respectively to C, Α, angle calculation module, rotating speed adaptive rate,

C is for output estimation stator current quadrature component under rotating coordinate system

with estimation stator current longitudinal axis component under rotating coordinate system

Wherein, a subtracter is used for quadrature component i under the actual stator electric current rotating coordinate system of input _sdwith quadrature component under estimation stator current rotating coordinate system

differ from, the error signal of the stator current of acquisition quadrature component under rotating coordinate system, and the error signal of this stator current quadrature component under rotating coordinate system is sent to rotating speed adaptive rate and G,

Another subtracter is used for longitudinal axis component i under the actual stator electric current rotating coordinate system of input _sqwith longitudinal axis component under estimation stator current rotating coordinate system

differ from, the error signal of the stator current of acquisition longitudinal axis component under rotating coordinate system, and the error signal of this stator current longitudinal axis component under rotating coordinate system is sent to rotating speed adaptive rate and G simultaneously,

Angle calculation module is used and rotor flux signal is carried out to angle calculation, and Α is used for exporting observation signal, and rotating speed adaptive rate is used for output speed feedback signal, and this speed feedback signal is sent to Α.

Claims (3)

1. the acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error, the method realizes based on existing full rank flux observer, it is characterized in that, and the method comprises the steps,

Step 1, below meeting, during 3 conditions, obtain 4 error feedback coefficients, and these 4 error feedback coefficients are respectively g ₁, g ₂, g ₃and g ₄, by 4 error feedback coefficient substitutions of obtaining

$G={\left[\begin{array}{cccc}{g}_1& {-g}_2& g_3& {-g}_4\\ g_2& g_1& g_4& g_3\end{array}\right]}^T$ (formula 1)

In, obtain G,

Wherein, G represents the Error Feedback matrix of observer,

3 conditions are respectively,

Condition one: observer limit real part is less than asynchronous machine limit real part, and be all negative,

Condition two: the zero limit real part of estimating rotating speed transfer function is all negative,

Condition three: utilize the error of estimating magnetic linkage and true magnetic linkage, assurance system, when motor low cruise, is equivalent to current model, and system, when high speed operation of motor, is equivalent to voltage model;

Step 2, according to known speed adaptive law equation:

${\widehat{\mathrm{\ω}}}_r=k_1(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})-k_2(e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})$ (formula 2),

Obtain the rear rotating speed adaptive law equation of distortion

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3),

Wherein, represent the motor speed of estimating,

K ₁represent stator current Error Gain,

E _{i α}expression estimation stator current is compared the error component of transverse axis under rest frame with actual stator electric current,

represent the longitudinal axis component of estimated rotor magnetic linkage under rest frame,

E _{i β}expression estimation stator current is compared the error component of the longitudinal axis under rest frame with actual stator electric current,

the quadrature component of statement estimated rotor magnetic linkage under rest frame,

K ₂represent rotor flux Error Gain,

E _{λ α}expression estimated rotor magnetic linkage is compared the error component of transverse axis under rest frame with actual rotor magnetic linkage,

represent the longitudinal axis component of estimated rotor magnetic linkage under rest frame,

E _{λ β}expression estimated rotor magnetic linkage is compared the error component of the longitudinal axis under rest frame with actual rotor magnetic linkage,

represent the quadrature component of estimated rotor magnetic linkage under rest frame,

K _prepresent the proportional gain of pi controller,

K _ithe storage gain that represents pi controller,

I _sqrepresent the longitudinal axis component of actual stator electric current under rotating coordinate system,

represent the longitudinal axis component of estimated rotor electric current under rotating coordinate system,

represent the quadrature component of estimated rotor magnetic linkage under rotating coordinate system,

K represents rotor flux amplitude error coefficient,

I _sdrepresent the quadrature component of actual stator electric current under rotating coordinate system,

represent to estimate the quadrature component of stator current under rotating coordinate system,

I _λrepresent rotor flux phase error coefficient,

Step 3, the Error Feedback matrix G of the observer obtaining in step 1 is replaced to the Error Feedback matrix in the flux observer of existing full rank, by rotating speed adaptive law equation after the distortion of obtaining in step 2

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3)

Replace the rotating speed adaptive law in the flux observer of existing full rank, successfully obtain the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error.

2. the acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error according to claim 1, is characterized in that, in described step 2, according to known speed adaptive law equation:

${\widehat{\mathrm{\ω}}}_r=k_1(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})-k_2(e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})$ (formula 2),

Obtain the rear rotating speed adaptive law equation of distortion

${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(i_{\mathrm{sq}}-{\hat{i}}_{\mathrm{sq}}){\widehat{\mathrm{\λ}}}_{\mathrm{rd}}+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}]$ (formula 3)

Detailed process be,

In (formula 2),

$e_{\mathrm{i\α}}=i_{\mathrm{s\α}}-{\hat{i}}_{\mathrm{s\α}}$ (formula 4),

$e_{\mathrm{i\β}}=i_{\mathrm{s\β}}-{\hat{i}}_{\mathrm{s\β}}$ (formula 5),

$e_{\mathrm{\λ\α}}={\mathrm{\λ}}_{\mathrm{r\α}}-{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}$ (formula 6),

$e_{\mathrm{\λ\β}}={\mathrm{\λ}}_{\mathrm{r\β}}-{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}$ (formula 7),

Wherein, i _{s α}represent the transverse axis stator current component actual value under static coordinate,

the estimated value that represents the transverse axis stator current component under static coordinate,

I _{s β}represent the longitudinal axis stator current component actual value under static coordinate,

the estimated value that represents the longitudinal axis stator current component under static coordinate,

λ _{r α}represent the transverse axis rotor flux component actual value under static coordinate,

represent the transverse axis rotor flux component estimated value under static coordinate,

λ _{r β}represent the longitudinal axis rotor flux component actual value under static coordinate,

represent the longitudinal axis rotor flux component estimated value under static coordinate,

By (formula 4) to (formula 7) substitution (formula 2)

in, carry out obtaining after abbreviation: $e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}={\mathrm{\λ}}_{\mathrm{r\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-{\mathrm{\λ}}_{\mathrm{r\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=\sqrt{{\mathrm{\λ}}_{\mathrm{r\α}}^2+{\mathrm{\λ}}_{\mathrm{r\β}}^2}\·\sqrt{{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}^2+{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}^2}\left(\frac{{\mathrm{\λ}}_{\mathrm{r\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-{\mathrm{\λ}}_{\mathrm{r\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}}{\sqrt{{\mathrm{\λ}}_{\mathrm{r\α}}^2+{\mathrm{\λ}}_{\mathrm{r\β}}^2}\·\sqrt{{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}^2+{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}^2}}\right)$ (formula 8),

Suppose under rest frame actual rotor flux linkage vector

with estimated rotor flux linkage vector

with the angle of α reference axis be respectively θ and

and θ and difference be △ θ, therefore, according to rotor flux Vector Rotation speed, equal stator current vector rotary speed, (formula 8) through distortion after obtain:

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=\left|{\stackrel{\→}{\mathrm{\λ}}}_r\right|\·\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|(\sin \widehat{\mathrm{\θ}}\cos \mathrm{\θ}-\cos \widehat{\mathrm{\θ}}\sin \mathrm{\θ})=-\sin \mathrm{\Δ\θ}({\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|}^2+\mathrm{\Δ\λ}|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\left|\right)$ (formula 9)

Wherein,

for the rotor flux amplitude of estimating,

for actual rotor flux amplitude, △ λ is the amplitude error of actual rotor magnetic linkage and estimated rotor magnetic linkage, and △ λ is 0,

Order

$k={\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|}^2+\mathrm{\Δ\λ}\left|{\stackrel{\widehat{\→}}{\mathrm{\λ}}}_r\right|$ (formula 10),

Wherein, k is rotor flux amplitude error coefficient,

Make the error of the rotor flux anglec of rotation of the actual rotor magnetic linkage anglec of rotation and estimation, equal the error of the stator current anglec of rotation of the actual stator current phasor anglec of rotation and observation, utilize the cosine law to obtain,

$\sin \mathrm{\Δ\θ}=\frac{|i_{\mathrm{sq}}{\hat{i}}_{\mathrm{sd}}-i_{\mathrm{sd}}{\hat{i}}_{\mathrm{sq}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 11),

Wherein, represent actual stator current phasor,

represent to estimate stator current vector amplitude,

(formula 10) and (formula 11) is updated in (formula 9), obtains

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=-k\frac{|i_{\mathrm{sq}}{\hat{i}}_{\mathrm{sd}}-i_{\mathrm{sd}}{\hat{i}}_{\mathrm{sq}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 12),

Will

be updated in (formula 12),

$e_{\mathrm{\λ\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{\λ\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}}=-k\frac{\left|i_{\mathrm{sq}}\right||i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}}|}{\left|{\stackrel{\→}{i}}_s\right|\·\left|{\stackrel{\widehat{\→}}{i}}_s\right|}$ (formula 13),

Utilize pi regulator (k _p+ k _i∫ dt) replace the k in (formula 2) ₁and k ₂, and (formula 13) is updated in (formula 2), obtain ${\widehat{\mathrm{\ω}}}_r=(k_p+k_i\∫\mathrm{dt})[(e_{\mathrm{i\α}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\β}}-e_{\mathrm{i\β}}{\widehat{\mathrm{\λ}}}_{\mathrm{r\α}})+k(i_{\mathrm{sd}}-{\hat{i}}_{\mathrm{sd}})i_{\mathrm{\λ}}].$

3. the acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor based on observation magnetic linkage error according to claim 1, it is characterized in that, described existing full rank flux observer comprises Α, B, C, G, 1/s, rotating speed adaptive rate, angle calculation module, an adder and two subtracters, described Α represents full rank flux observation matrix, B represents voltage input matrix, C represents electric current output matrix, and 1/s represents integral operation

Adder, for the error compensating signal of the observation signal of B output voltage signal, Α output and G output is sued for peace, obtains rotor flux differential signal,

1/s, for the rotor flux differential signal of adder output is carried out to integral operation, obtains rotor flux signal, and rotor flux signal is sent to respectively to C, Α, angle calculation module, rotating speed adaptive rate,

C is for output estimation stator current quadrature component under rotating coordinate system

with estimation stator current longitudinal axis component under rotating coordinate system

Wherein, a subtracter is used for quadrature component i under the actual stator electric current rotating coordinate system of input _sdwith quadrature component under estimation stator current rotating coordinate system

differ from, the error signal of the stator current of acquisition quadrature component under rotating coordinate system, and the error signal of this stator current quadrature component under rotating coordinate system is sent to rotating speed adaptive rate and G,

Another subtracter is used for longitudinal axis component i under the actual stator electric current rotating coordinate system of input _sqwith longitudinal axis component under estimation stator current rotating coordinate system

differ from, the error signal of the stator current of acquisition longitudinal axis component under rotating coordinate system, and the error signal of this stator current longitudinal axis component under rotating coordinate system is sent to rotating speed adaptive rate and G simultaneously,

Angle calculation module is used and rotor flux signal is carried out to angle calculation, and Α is used for exporting observation signal, and rotating speed adaptive rate is used for output speed feedback signal, and this speed feedback signal is sent to Α.

Images