CN103904973B - A kind of method realizing salient pole permanent magnet synchronous motor senseless control - Google Patents

Abstract

The invention discloses a kind of salient pole permanent magnet synchronous motor Speed Sensorless Control Method, belong to electric machine speed regulation field.It is characterized in that, according to the stator voltage recorded in real time and stator current, estimate to obtain effective back electromotive force by the full scalariform state sliding mode observer set up, then obtain rotor-position and rotating speed by phase-locked loop.The invention solves the problem of salient pole permanent magnet synchronous motor senseless control, senseless control is made to be subject to motor parameter influence less, and when realizing rotor-position and rotating speed estimates simultaneously, improve the robustness of rotor position estimate to speed estimate deviation, it also avoid magnifying slip mode noise, ensure that the stability of system.

Description

A kind of method realizing salient pole permanent magnet synchronous motor senseless control

Technical field

The present invention relates to a kind of method of motor senseless control, in particular for a kind of method realizing salient pole permanent magnet synchronous motor senseless control of wind power generation occasion.

Background technology

In order to tackle energy crisis, new energy technology obtains vigorous growth.Wind energy is a kind of inexhaustible, nexhaustible regenerative resource, and wind generating technology has become the most important thing of new energy technology.Salient pole permanent magnet synchronous motor is widely used in wind generator system because of the advantage such as high power density, high reliability.For improving system reliability, wind generator system usually requires to realize senseless control.Conventional Speed Sensorless Control Method has a lot, as High Frequency Injection, model reference adaptive method and sliding mode observer method etc.

Application for a patent for invention " a kind of position sensor-free vector control device for built-in permanent magnetic synchronous motor " (CN102361430A) and " motor without position sensor control system and control method " (CN102624322A) are proposed the permagnetic synchronous motor Speed Sensorless Control Method based on high frequency electrocardiography, but High Frequency Injection is only applicable to low-speed region, be not suitable for high-power wind power generation system.Application for a patent for invention " a kind of permagnetic synchronous motor method for controlling position-less sensor " (CN103051271A) devises the permagnetic synchronous motor Speed Sensorless Control Method based on model reference adaptive principle, but more serious by the impact of the parameter of electric machine.Application for a patent for invention " sensorless control system of permagnetic synchronous motor " (CN101964624A) proposes the permagnetic synchronous motor senseless control algorithm based on Second Order Sliding Mode observer, but, because the back electromotive force estimated contains high frequency sliding formwork noise, filtering must be carried out, filtering but brings phase delay, also must carry out phase compensation, cause system to realize complicated, angle estimation affects seriously by speed estimate value.

The meeting paper " RotorPositionEstimationwithFull-OrderSliding-ModeObserve rforSensorlesssIPMSM " that Harbin Institute of Technology king Gao Lin in 2012 etc. deliver proposes the salient pole permanent magnet synchronous motor Speed Sensorless Control Method based on expansion back electromotive force, solve Second Order Sliding Mode Control Problems existing, but based on expanding the impact of salient pole permanent magnet synchronous motor model by motor stator resistance, d axle inductance and q axle inductance three parameters of back electromotive force.The paper " AFamilyofSensorlessObserverswithSpeedEstimateforRotorPos itionEstimationofIMandPMSMDrives " that the scholars such as MihaiComanescu in 2012 deliver devises a full-order sliding mode observer for face dress formula permagnetic synchronous motor, have studied the robustness of rotor position estimate to error for rotating speed estimation, but in order to improve its robustness, need to increase sliding formwork gain, increase sliding formwork gain meeting magnifying slip mode noise, system can be caused time serious unstable, and the method can not directly apply in salient pole permanent magnet synchronous motor.

Summary of the invention

The technical problem to be solved in the present invention is the limitation overcoming above-mentioned various technical scheme, provides a kind of a kind of method realizing salient pole permanent magnet synchronous motor senseless control state observer and sliding mode observer combined together.

For solving technical problem of the present invention, the technical scheme adopted is: a kind of method realizing salient pole permanent magnet synchronous motor senseless control comprises stator voltage u _aB, u _bC, u _cAwith stator current i _a, i _b, i _cmeasurement, particularly,

According to the stator voltage u recorded in real time _aB, u _bC, u _cAwith stator current i _a, i _b, i _c, the effective back electromotive force estimated is obtained by full scalariform state sliding mode observer the effective back electromotive force estimated rotor-position and rotating speed is obtained again by phase-locked loop;

Described full scalariform state sliding mode observer is:

$\left\{\begin{array}{c}p{\hat{i}}_{\mathrm{\α}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\α}}+\frac{u_{\mathrm{\α}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\α}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\α}}\\ p{\hat{i}}_{\mathrm{\β}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\β}}+\frac{u_{\mathrm{\β}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\β}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\β}}\\ p{\hat{e}}_{\mathrm{\α}}=-{\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}+{\mathrm{Ns}}_{\mathrm{\β}}\\ p{\hat{e}}_{\mathrm{\β}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}+{\mathrm{Ns}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, $s_{\mathrm{\α}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}})+K({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}}),s_{\mathrm{\β}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}})+K({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}}),$ M is sliding formwork gain, sgn() be sign function, K is state gain, and N is effective back electromotive force e _α, e _βthe gain of sliding mode observer, for the estimation current value on static alpha-beta coordinate system, for the effective back electromotive force estimated, for the synchronous speed estimated, L _qfor motor q axle inductance, R _sfor motor stator resistance.

As a kind of further improvements in methods realizing salient pole permanent magnet synchronous motor senseless control:

Preferably, the effective back electromotive force estimated is obtained step as follows:

Step 1, if motor stator winding is delta connection, then performs step 1.1, if motor stator winding is star connection, then performs step 1.2,

Step 1.1, first sample motor stator voltage u _aB, u _bC, u _cA, then by u _aB, u _bC, u _cAcoordinate transform on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{u}_{\mathrm{\α}}\\ u_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{u}_{\mathrm{AB}}\\ u_{\mathrm{BC}}\\ u_{\mathrm{CA}}\end{array}\right],$

Wherein, u _α, u _βfor the component of stator voltage on static alpha-beta coordinate system, afterwards, first sample motor stator current i _a, i _b, then calculate phase current, as shown in the formula:

$\left\{\begin{array}{c}{i}_{\mathrm{Ax}=1/3(i_A-i_B)}\\ i_{\mathrm{Bx}=i_{\mathrm{Ax}}+i_B}\\ i_{\mathrm{Cx}}=i_{\mathrm{Ax}}-i_A\end{array}\right.$

Wherein, i _ax, i _bx, i _cxfor three-phase phase current, finally, by the coordinate transform of three-phase phase current on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{i}_{\mathrm{\α}}\\ i_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{Ax}}\\ i_{\mathrm{Bx}}\\ i_{\mathrm{Cx}}\end{array}\right],$

Wherein, i _α, i _βfor the component of phase current on static alpha-beta coordinate system;

Step 1.2, first sample motor stator voltage u _aB, u _bC, then calculate machine phase voltages, as shown in the formula:

$\left\{\begin{array}{c}{u}_{B=1/3(u_{\mathrm{BC}}-u_{\mathrm{AB}})}\\ u_{A=u_B+u_{\mathrm{AB}}}\\ u_C=u_B-u_{\mathrm{BC}}\end{array}\right.,$

Wherein, u _a, u _b, u _cfor three-phase phase voltage, afterwards, by three-phase phase voltage coordinate transform on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{u}_{\mathrm{\α}}\\ u_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{u}_A\\ u_B\\ u_C\end{array}\right],$

Wherein, u _α, u _βfor the component of stator voltage on static alpha-beta coordinate system, finally, first sample motor stator current i _a, i _b, i _c, then obtain the component of electric current on static alpha-beta coordinate system through coordinate transform, as shown in the formula:

$\left[\begin{array}{c}{i}_{\mathrm{\α}}\\ i_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{i}_A\\ i_B\\ i_C\end{array}\right],$

Wherein, i _α, i _βfor the component of phase current on static alpha-beta coordinate system;

Step 2, first with the electric current estimated deduct the actual current i that step 1 obtains _α, i _β, obtain current deviation again respectively through sign function sgn() computing and multiplying, finally obtain current error feedback quantity s _α, s _β, as shown in the formula:

$s_{\mathrm{\α}}=\mathrm{Msgn}\left({\stackrel{\‾}{i}}_{\mathrm{\α}}\right)+K\left({\stackrel{\‾}{i}}_{\mathrm{\α}}\right),s_{\mathrm{\β}}=\mathrm{Msgn}\left({\stackrel{\‾}{i}}_{\mathrm{\β}}\right)+K\left({\stackrel{\‾}{i}}_{\mathrm{\β}}\right),$

Wherein, M is sliding formwork gain, and K is state gain,

Step 3, the voltage u obtained by step 1 _α, u _βdeduct the current error feedback quantity s that step 2 obtains _α, s _β, then deduct effective back electromotive force of the estimation that kth time computing obtains obtain first group of intermediate variable E _{1 α}, E _{1 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{1\mathrm{\α}}=u_{\mathrm{\α}}-s_{\mathrm{\α}}-{\hat{e}}_{\mathrm{\α}}\left(k\right)\\ E_{1\mathrm{\β}}=u_{\mathrm{\β}}-s_{\mathrm{\β}}-{\hat{e}}_{\mathrm{\β}}\left(k\right)\end{array}\right.,$

Wherein, E _{1 α}, E _{1 β}be first group of intermediate variable, k is positive integer, (k=0,1,2), ${\hat{e}}_{\mathrm{\β}}\left(0\right)=0;$

Step 4, the first group of intermediate variable E obtained by step 3 _{1 α}, E _{1 β}divided by motor q axle inductance, obtain second group of intermediate variable E _{2 α}, E _{2 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{2\mathrm{\α}}=E_{1\mathrm{\α}}/L_q\\ E_{2\mathrm{\β}}=E_{1\mathrm{\β}}/L_q\end{array}\right.,$

Wherein, L _qfor motor q axle inductance, E _{2 α}, E _{2 β}be second group of intermediate variable;

Step 5, according to the motor stator electric current estimated with motor stator resistance R _sand q axle inductance L _q, calculate the 3rd group of intermediate variable E _{3 α}, E _{3 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{3\mathrm{\α}}=\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\α}}\\ E_{3\mathrm{\β}}=\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, E _{3 α}, E _{3 β}be the 3rd group of intermediate variable, Rs is motor stator resistance;

Step 6, the E obtained by step 4 _{2 α}, E _{2 β}deduct the E that step 5 obtains _{3 α}, E _{3 β}, obtain the 4th group of intermediate variable E _{4 α}, E _{4 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{4\mathrm{\α}}=E_{2\mathrm{\α}}-E_{3\mathrm{\α}}\\ E_{4\mathrm{\β}}=E_{2\mathrm{\β}}-E_{3\mathrm{\β}}\end{array}\right.,$

Wherein, E _{4 α}, E _{4 β}be the 4th group of intermediate variable;

Step 7, to the E that step 6 obtains _{4 α}, E _{4 β}carry out integration, obtain the electric current estimated as shown in the formula:

$\left\{\begin{array}{c}{\hat{i}}_{\mathrm{\α}}={\hat{i}}_{\mathrm{\α}}+T_s{E}_{4\mathrm{\α}}\\ {\hat{i}}_{\mathrm{\β}}={\hat{i}}_{\mathrm{\β}}+T_s{E}_{4\mathrm{\β}}\end{array}\right.,$

Wherein, T _sfor the sampling period;

Step 8, the current estimation value that step 7 is obtained bring step 2 and step 5 into, for estimating stator current further;

Step 9, the current error feedback quantity s that step 2 is obtained _α, s _βbe multiplied by gain N, obtain the 5th group of intermediate variable E _{5 α}, E _{5 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{5\mathrm{\α}}={\mathrm{Ns}}_{\mathrm{\α}}\\ E_{5\mathrm{\β}}={\mathrm{Ns}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, E _{5 α}, E _{5 β}be the 5th group of intermediate variable, N is effective back electromotive force e _α, e _βthe gain of sliding mode observer;

Step 10, the rotating speed of estimation be multiplied by effective back electromotive force of the estimation that kth time computing obtains respectively obtain the 6th group of intermediate variable E _{6 α}, E _{6 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{6\mathrm{\α}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}\left(k\right)\\ E_{6\mathrm{\β}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\β}}\left(k\right)\end{array}\right.,$

Wherein, E _{6 α}, E _{6 β}be the 6th group of intermediate variable;

Step 11, the E first obtained by step 9 _{5 α}deduct the E that step 10 obtains _{6 β}, obtain the 7th intermediate variable E _{7 α}, then the E obtained by step 9 _{5 β}add the E that step 10 obtains _{6 α}, obtain the 7th intermediate variable E _{7 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{7\mathrm{\α}}=E_{5\mathrm{\α}}-E_{6\mathrm{\β}}\\ E_{7\mathrm{\β}}=E_{5\mathrm{\β}}+E_{6\mathrm{\α}}\end{array}\right.,$

Wherein, E _{7 α}, E _{7 β}be the 7th group of intermediate variable;

Step 12, respectively to the E that step 11 obtains _{7 α}, E _{7 β}carry out integration, obtain effective back electromotive force that kth+1 computing is estimated as shown in the formula:

$\left\{\begin{array}{c}{\hat{e}}_{\mathrm{\α}}(k+1)={\hat{e}}_{\mathrm{\α}}\left(k\right)+T_s{E}_{7\mathrm{\α}}\\ {\hat{e}}_{\mathrm{\β}}(k+1)={\hat{e}}_{\mathrm{\β}}\left(k\right)+T_s{E}_{7\mathrm{\β}}\end{array}\right.;$

Step 13, effective back electromotive force of the estimation first step 12 obtained bring step 3 into, be further used for solving first group of intermediate variable E _{1 α}, E _{1 β}, more effective back electromotive force of the estimation that step 12 is obtained bring step 10 into, for calculating the 7th group of intermediate variable E _{7 α}, E _{7 β};

Step 14, effective back electromotive force of the estimation that step 12 is obtained bring phase-locked loop into, obtain the rotor-position estimated with the rotating speed estimated

Step 15, by the rotating speed estimated bring step 10 into, for calculating the 7th group of intermediate variable E _{7 α}, E _{7 β};

Step 16, repeats step 1 ~ step 15, until the current value estimated equal actual current value i _α, i _β, obtain the back electromotive force e with reality _α, e _βeffective back electromotive force of the estimation conformed to and the rotor-position of the estimation to conform to the rotor position of reality

Preferably, phase-locked loop is according to the effective back electromotive force estimated obtain rotor-position with the rotating speed estimated step as follows:

Step 1, according to the effective back electromotive force of estimation calculate effective back electromotive force q axle deviation, as shown in the formula:

${\hat{e}}_q=-{\hat{e}}_{\mathrm{\α}}(k+1)\cos \widehat{\mathrm{\θ}}-{\hat{e}}_{\mathrm{\β}}(k+1)\sin \widehat{\mathrm{\θ}},$

Wherein, for effective back electromotive force q axle deviation;

Step 2, is obtained by step 1 the rotating speed estimated is obtained through proportional and integral controller as shown in the formula:

${\widehat{\mathrm{\ω}}}_r=(k_i/s+k_p){\hat{e}}_q,$

Wherein, k _pand k _ibe respectively ratio and integral coefficient, s is Laplacian;

Step 3, the estimation rotating speed obtained by step 2 the rotor-position estimated is obtained by integration as shown in the formula:

$\widehat{\mathrm{\θ}}=\widehat{\mathrm{\θ}}+T_s{\widehat{\mathrm{\ω}}}_r.$

Relative to the beneficial effect of prior art be:

The present invention is based on the full scalariform state sliding mode observer of the conceptual design of effective back electromotive force one.The traditional full rank of the salient pole permanent magnet synchronous motor based on effective magnetic linkage model is as follows:

$\left\{\begin{array}{c}{\mathrm{pi}}_{\mathrm{\α}}=\frac{R_s}{L_q}{i}_{\mathrm{\α}}+\frac{u_{\mathrm{\α}}}{L_q}-\frac{{\mathrm{p\ψ}}_{\mathrm{\α}}}{L_q}\\ {\mathrm{pi}}_{\mathrm{\β}}=-\frac{R_s}{L_q}{i}_{\mathrm{\β}}+\frac{u_{\mathrm{\β}}}{L_q}-\frac{{\mathrm{p\ψ}}_{\mathrm{\β}}}{L_q}\\ {\mathrm{p\ψ}}_{\mathrm{\α}}=-{\mathrm{\ω}}_r{\mathrm{\ψ}}_{\mathrm{\β}}\\ {\mathrm{p\ψ}}_{\mathrm{\β}}={\mathrm{\ω}}_r{\mathrm{\ψ}}_{\mathrm{\α}}\end{array}\right.$

$\left[\begin{array}{c}{\mathrm{\ψ}}_{\mathrm{\β}}\\ {\mathrm{\ψ}}_{\mathrm{\α}}\end{array}\right]=[{\mathrm{\ψ}}_f+(L_d-L_q)i_d]\left[\begin{array}{c}{\sin \mathrm{\θ}}_r\\ \cos {\mathrm{\θ}}_r\end{array}\right];$

Wherein, i _α, i _βbe respectively two components of stator current on static alpha-beta coordinate system, u _α, u _βbe respectively two components of stator voltage on static alpha-beta coordinate system, ψ _α, ψ _βfor two components of effective magnetic linkage on static alpha-beta coordinate system, ψ _ffor permanent magnet flux linkage, L _dfor d axle inductance, L _qfor q axle inductance, R _sfor stator resistance, ω _rfor synchronous speed, θ _rfor rotor-position.

Effective back electromotive force e of the present invention's definition _α, e _βas follows:

$\left\{\begin{array}{c}{e}_{\mathrm{\α}}={\mathrm{p\ψ}}_{\mathrm{\α}}\\ e_{\mathrm{\β}}={\mathrm{p\ψ}}_{\mathrm{\β}}\end{array}\right..$

Think that magnetic linkage amplitude is constant to obtain:

$\left\{\begin{array}{c}{\mathrm{pe}}_{\mathrm{\α}}=-{\mathrm{\ω}}_r{e}_{\mathrm{\β}}\\ {\mathrm{pe}}_{\mathrm{\β}}={\mathrm{\ω}}_r{e}_{\mathrm{\α}}\end{array}\right..$

According to effective back electromotive force e _α, e _βdefinition, the full rank of the salient pole permanent magnet synchronous motor after being improved model is as follows:

$\left\{\begin{array}{c}{\mathrm{pi}}_{\mathrm{\α}}=-\frac{R_s}{L_q}{i}_{\mathrm{\α}}+\frac{u_{\mathrm{\α}}}{L_q}-\frac{e_{\mathrm{\α}}}{L_q}\\ {\mathrm{pi}}_{\mathrm{\β}}=-\frac{R_s}{L_q}{i}_{\mathrm{\β}}+\frac{u_{\mathrm{\β}}}{L_q}-\frac{e_{\mathrm{\β}}}{L_q}\\ {\mathrm{pe}}_{\mathrm{\α}}=-{\mathrm{\ω}}_r{e}_{\mathrm{\β}}\\ {\mathrm{pe}}_{\mathrm{\β}}={\mathrm{\ω}}_r{e}_{\mathrm{\α}}\end{array}\right..$

The paper " AFamilyofSensorlessObserverswithSpeedEstimateforRotorPos itionEstimationofIMandPMSMDrives " that the scholars such as MihaiComanescu in 2012 deliver devises a full-order sliding mode observer for face dress formula permagnetic synchronous motor, and its form is consistent with above formula.But also there is the problem that sliding formwork is buffeted in full-order sliding mode observer.For this reason, based on the modified model salient pole permanent magnet synchronous motor full rank model of the present invention's design, devise full scalariform state sliding mode observer, introduce state error item, sliding formwork can be weakened and buffet, improve the stability of a system.

The full scalariform state sliding mode observer set up based on this model is as follows:

$\left\{\begin{array}{c}p{\hat{i}}_{\mathrm{\α}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\α}}+\frac{u_{\mathrm{\α}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\α}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\α}}\\ p{\hat{i}}_{\mathrm{\β}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\β}}+\frac{u_{\mathrm{\β}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\β}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\β}}\\ p{\hat{e}}_{\mathrm{\α}}=-{\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\β}}+{\mathrm{Ns}}_{\mathrm{\α}}\\ p{\hat{e}}_{\mathrm{\β}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}+{\mathrm{Ns}}_{\mathrm{\β}}\end{array}\right.;$

Wherein, $s_{\mathrm{\α}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}})+K({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}}),s_{\mathrm{\β}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}})+K({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}}),$ M is sliding formwork gain, sgn() be sign function, K is state gain, and N is effective back electromotive force e _α, e _βthe gain of sliding mode observer, for the estimation current value on static alpha-beta coordinate system, for the effective back electromotive force estimated, for the synchronous speed estimated.

According to the full scalariform state sliding mode observer that the present invention sets up, effective back electromotive force can be realized estimation, then can rotating speed be obtained by phase-locked loop and rotor-position

The present invention in order to by full-order sliding mode observer algorithm application in salient pole permanent magnet synchronous motor, first improve the salient pole permanent magnet synchronous motor model based on effective magnetic linkage, propose the concept of effective back electromotive force; Secondly state observer and sliding mode observer are combined together, devise state sliding mode observer, solve when improving the robustness of rotor position estimate to error for rotating speed estimation and need to increase sliding formwork gain, and suppress to need during sliding formwork noise to reduce this conflict of sliding formwork gain, achieve both making overall plans, thus improve stability and the robustness of system.Through emulation actual measurement, the present invention can realize the accurate estimation of rotor-position automatically, realizes stable senseless control.

The invention solves the problem of salient pole permanent magnet synchronous motor senseless control, senseless control is made to be subject to motor parameter influence less, and when realizing rotor-position and rotating speed estimates simultaneously, improve the robustness of rotor position estimate to speed estimate deviation, it also avoid magnifying slip mode noise, ensure that the stability of system.

Accompanying drawing explanation

Below in conjunction with accompanying drawing, optimal way of the present invention is described in further detail.

Fig. 1 is the concrete enforcement block diagram of current status sliding mode observer.

Fig. 2 is the concrete enforcement block diagram of effective back electromotive force state sliding mode observer.

Fig. 3 is phase-locked loop specifically enforcement figure.

Simulation waveform figure when rotating speed has a deviation is fed back when Fig. 4 is M=100, N=50.

Simulation waveform figure when rotating speed has a deviation is fed back when Fig. 5 is M=100, N=500.

Embodiment

See Fig. 1, Fig. 2, Fig. 3, Fig. 4 and Fig. 5, a kind of method realizing salient pole permanent magnet synchronous motor senseless control is as follows:

First stator voltage u is carried out _aB, u _bC, u _cAwith stator current i _a, i _b, i _cmeasurement, then according to the stator voltage u that records in real time _aB, u _bC, u _cAwith stator current i _a, i _b, i _c, the effective back electromotive force estimated is obtained by full scalariform state sliding mode observer the effective back electromotive force estimated rotor-position and rotating speed is obtained again by phase-locked loop;

Described full scalariform state sliding mode observer is:

$\left\{\begin{array}{c}p{\hat{i}}_{\mathrm{\α}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\α}}+\frac{u_{\mathrm{\α}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\α}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\α}}\\ p{\hat{i}}_{\mathrm{\β}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\β}}+\frac{u_{\mathrm{\β}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\β}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\β}}\\ p{\hat{e}}_{\mathrm{\α}}=-{\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\β}}+{\mathrm{Ns}}_{\mathrm{\α}}\\ p{\hat{e}}_{\mathrm{\β}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}+{\mathrm{Ns}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, $s_{\mathrm{\α}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}})+K({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}}),s_{\mathrm{\β}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}})+K({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}}),$ M is sliding formwork gain, sgn() be sign function, K is state gain, and N is effective back electromotive force e _α, e _βthe gain of sliding mode observer, for the estimation current value on static alpha-beta coordinate system, for the effective back electromotive force estimated, for the synchronous speed estimated, L _qfor motor q axle inductance, R _sfor motor stator resistance.

Wherein, concrete steps of the present invention are as follows:

As shown in Figure 1, the component u of stator voltage on static alpha-beta coordinate system first sampled needed for calculating observation device _α, u _βwith the component i of phase current on static alpha-beta coordinate system _α, i _β, more in two kinds of situation: if motor stator winding is delta connection, then perform step (a), if motor stator winding is star connection, then perform step (b);

Step (a), sample motor stator voltage u _aB, u _bC, u _cA, then by u _aB, u _bC, u _cAcoordinate transform is on static alpha-beta coordinate system, as follows:

$\left[\begin{array}{c}{u}_{\mathrm{\α}}\\ u_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{u}_{\mathrm{AB}}\\ u_{\mathrm{BC}}\\ u_{\mathrm{CA}}\end{array}\right],$

Wherein, u _α, u _βfor the component of stator voltage on static alpha-beta coordinate system,

Sample motor stator current i _a, i _b, then calculate phase current, as shown in the formula:

$\left\{\begin{array}{c}{i}_{\mathrm{Ax}=1/3(i_A-i_B)}\\ i_{\mathrm{Bx}=i_{\mathrm{Ax}}+i_B}\\ i_{\mathrm{Cx}}=i_{\mathrm{Ax}}-i_A\end{array}\right.,$

Wherein, i _ax, i _bx, i _cxfor three-phase phase current,

Then by the coordinate transform of three-phase phase current on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{i}_{\mathrm{\α}}\\ i_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{i}_{\mathrm{Ax}}\\ i_{\mathrm{Bx}}\\ i_{\mathrm{Cx}}\end{array}\right],$

Wherein, i _α, i _βfor the component of phase current on static alpha-beta coordinate system;

Step (b), sample motor stator voltage u _aB, u _bC, then calculate machine phase voltages, as shown in the formula:

$\left\{\begin{array}{c}{u}_{B=1/3(u_{\mathrm{BC}}-u_{\mathrm{AB}})}\\ u_{A=u_B+u_{\mathrm{AB}}}\\ u_C=u_B-u_{\mathrm{BC}}\end{array}\right.,$

Wherein, u _a, u _b, u _cfor three-phase phase voltage,

Then by three-phase phase voltage coordinate transform on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{u}_{\mathrm{\α}}\\ u_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{u}_A\\ u_B\\ u_C\end{array}\right],$

Wherein, u _α, u _βfor the component of stator voltage on static alpha-beta coordinate system,

Sample motor stator current i _a, i _b, i _c, then obtain the component of electric current on static alpha-beta coordinate system through coordinate transform, as shown in the formula:

$\left[\begin{array}{c}{i}_{\mathrm{\α}}\\ i_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{i}_A\\ i_B\\ i_C\end{array}\right],$

Wherein, i _α, i _βfor the component of phase current on static alpha-beta coordinate system;

Obtaining voltage u _α, u _βand current i _α, i _βafter, as shown in Figure 1, then according to voltage u _α, u _β, effective back electromotive force of estimation of obtaining of kth time computing with current error feedback quantity s _α, s _β, obtain first group of intermediate variable E _{1 α}, E _{1 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{1\mathrm{\α}}=u_{\mathrm{\α}}-s_{\mathrm{\α}}-{\hat{e}}_{\mathrm{\α}}\left(k\right)\\ E_{1\mathrm{\β}}=u_{\mathrm{\β}}-s_{\mathrm{\β}}-{\hat{e}}_{\mathrm{\β}}\left(k\right)\end{array}\right.,$

Wherein, E _{1 α}, E _{1 β}be first group of intermediate variable, s _α, s _βfor current error feedback quantity, k is positive integer, (k=0,1,2),

S _α, s _βbe calculated as follows:

$s_{\mathrm{\α}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}})+K({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}}),s_{\mathrm{\β}}=\mathrm{Msgn}({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}})+K({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}}),$

Wherein, M is sliding formwork gain, and K is state gain, the electric current estimated;

As shown in Figure 1, according to one group of intermediate variable E _{1 α}, E _{1 β}calculate second group of intermediate variable E _{2 α}, E _{2 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{2\mathrm{\α}}=E_{1\mathrm{\α}}/L_q\\ E_{2\mathrm{\β}}=E_{1\mathrm{\β}}/L_q\end{array}\right.,$

Wherein, L _qfor motor q axle inductance, E _{2 α}, E _{2 β}be second group of intermediate variable;

As shown in Figure 1, according to second group of intermediate variable E _{2 α}, E _{2 β}calculate the 4th group of intermediate variable E _{4 α}, E _{4 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{4\mathrm{\α}}=E_{2\mathrm{\α}}-E_{3\mathrm{\α}}\\ E_{4\mathrm{\β}}=E_{2\mathrm{\β}}-E_{3\mathrm{\β}}\end{array}\right.,$

Wherein, E _{3 α}, E _{3 β}be the 3rd group of intermediate variable, E _{4 α}, E _{4 β}be the 4th group of intermediate variable; 3rd group of intermediate variable E _{3 α}, E _{3 β}be calculated as follows formula:

$\left\{\begin{array}{c}{E}_{3\mathrm{\α}}=\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\α}}\\ E_{3\mathrm{\β}}=\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, R _sfor motor stator resistance;

As shown in Figure 1, to E _{4 α}, E _{4 β}carry out integration, obtain the electric current estimated as shown in the formula:

$\left\{\begin{array}{c}{\hat{i}}_{\mathrm{\α}}={\hat{i}}_{\mathrm{\α}}+T_s{E}_{4\mathrm{\α}}\\ {\hat{i}}_{\mathrm{\β}}={\hat{i}}_{\mathrm{\β}}+T_s{E}_{4\mathrm{\β}}\end{array}\right.,$

Wherein, T _sfor the sampling period;

As shown in Figure 2, current error feedback quantity s _α, s _βbe multiplied by gain N and obtain the 5th group of intermediate variable E _{5 α}, E _{5 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{5\mathrm{\α}}={\mathrm{Ns}}_{\mathrm{\α}}\\ E_{5\mathrm{\β}}={\mathrm{Ns}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, E _{5 α}, E _{5 β}be the 5th group of intermediate variable, N is effective back electromotive force e _α, e _βthe gain of sliding mode observer;

As shown in Figure 2, with the rotating speed estimated be multiplied by effective back electromotive force of the estimation that kth time computing obtains respectively obtain the 6th group of intermediate variable E _{6 α}, E _{6 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{6\mathrm{\α}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}\left(k\right)\\ E_{6\mathrm{\β}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\β}}\left(k\right)\end{array}\right.,$

Wherein, E _{6 α}, E _{6 β}be the 6th group of intermediate variable;

As shown in Figure 2, E _{5 α}deduct E _{6 β}, obtain the 7th intermediate variable E _{7 α}, E _{5 β}add E _{6 α}, obtain the 7th intermediate variable E _{7 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{7\mathrm{\α}}=E_{5\mathrm{\α}}-E_{6\mathrm{\β}}\\ E_{7\mathrm{\β}}=E_{5\mathrm{\β}}+E_{6\mathrm{\α}}\end{array}\right.,$

Wherein, E _{7 α}, E _{7 β}be the 7th group of intermediate variable;

As shown in Figure 2, to E _{7 α}, E _{7 β}carry out integration respectively, obtain effective back electromotive force that kth+1 computing is estimated ${\hat{e}}_{\mathrm{\α}}(k+1){,\hat{e}}_{\mathrm{\β}}(k+1),$ As shown in the formula:

$\left\{\begin{array}{c}{\hat{e}}_{\mathrm{\α}}(k+1)={\hat{e}}_{\mathrm{\α}}\left(k\right)+T_s{E}_{7\mathrm{\α}}\\ {\hat{e}}_{\mathrm{\β}}(k+1)={\hat{e}}_{\mathrm{\β}}\left(k\right)+T_s{E}_{7\mathrm{\β}}\end{array}\right.;$

As shown in Figure 3, according to effective back electromotive force that kth+1 computing is estimated calculate effective back electromotive force q axle deviation, as shown in the formula:

${\hat{e}}_q=-{\hat{e}}_{\mathrm{\α}}(k+1)\cos \widehat{\mathrm{\θ}}-{\hat{e}}_{\mathrm{\β}}(k+1)\sin \widehat{\mathrm{\θ}},$

Wherein, for effective back electromotive force q axle deviation;

As shown in Figure 3, the rotating speed estimated is obtained through proportional and integral controller as shown in the formula:

${\widehat{\mathrm{\ω}}}_r=(k_i/s+k_p){\hat{e}}_q,$

Wherein, k _pand k _ibe respectively ratio and integral coefficient, s is Laplacian;

As shown in Figure 3, the rotating speed of estimation the rotor-position estimated is obtained by integration as shown in the formula:

$\widehat{\mathrm{\θ}}=\widehat{\mathrm{\θ}}+T_s{\widehat{\mathrm{\ω}}}_r.$

In order to verify validity of the present invention, carry out simulating, verifying.Motor stator resistance used is 0.34 Europe, and d axle inductance is 0.046H, q axle inductance is 0.135H, and permanent magnet flux linkage is 0.75Wb.The full scalariform state sliding mode observer adopting this method to propose and phase-locked loop realize rotor position estimate.During emulation, given exciting current i _sd=0, i _sq=10A, given motor speed is 60Rad/s.First from the full scalariform state sliding mode observer that this method is set up, because full scalariform state sliding mode observer is not containing d axle inductance, therefore not by the impact of d axle inductance.In order to verify the robustness of the method rotor position estimate to error for rotating speed estimation further, during emulation, on feedback speed, an error coefficient 0.8 is artificially multiplied by when 0.6s, namely adopt the true velocity estimated as feedback before 0.6s, the speed of mistake is adopted to feed back after 0.6s, design M=1, N=50, during K=50, simulation result as shown in Figure 4, it is visible when deviation appears in feedback speed, estimate that deviation has also appearred in angle, during further increase N=500, simulation result as shown in Figure 5, visible, by increasing sliding formwork gain N, the robustness of rotor position estimate to error for rotating speed estimation can be improved.If select K=0, then the full scalariform state sliding mode observer abbreviation of this method design is full-order sliding mode observer, now for ensureing system stability, must be when increasing M, and in order to keep sliding formwork gain constant, namely keep MN constant, then must reduce N, therefore, rotor position estimate also can reduce the robustness of error for rotating speed estimation.The advantage of state sliding mode observer is just, by introducing state error feedback factor K, can allow to reduce sliding formwork gain M, thus allowing in certain sliding formwork noise range, improve sliding formwork gain N, to improve the robustness of rotor position estimate to error for rotating speed estimation further.

Obviously, those skilled in the art can carry out various change and modification to a kind of method realizing salient pole permanent magnet synchronous motor senseless control of the present invention and not depart from the spirit and scope of the present invention.Like this, if belong within the scope of the claims in the present invention and equivalent technologies thereof to these amendments of the present invention and modification, then the present invention is also intended to comprise these change and modification.

Claims (2)

1. realize a method for salient pole permanent magnet synchronous motor senseless control, comprise stator voltage u _aB, u _bC, u _cAwith stator current i _a, i _b, i _cmeasurement, it is characterized in that:

According to the stator voltage u recorded in real time _aB, u _bC, u _cAwith stator current i _a, i _b, i _c, the effective back electromotive force estimated is obtained by full scalariform state sliding mode observer the effective back electromotive force estimated rotor-position and rotating speed is obtained again by phase-locked loop; Wherein,

(1) full scalariform state sliding mode observer is:

$\left\{\begin{array}{c}p{\hat{i}}_{\mathrm{\α}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\α}}+\frac{u_{\mathrm{\α}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\α}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\α}}\\ p{\hat{i}}_{\mathrm{\β}}=-\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\β}}+\frac{u_{\mathrm{\β}}}{L_q}-\frac{{\hat{e}}_{\mathrm{\β}}}{L_q}-\frac{1}{L_q}{s}_{\mathrm{\β}}\\ p{\hat{e}}_{\mathrm{\α}}=-{\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\β}}+{\mathrm{Ns}}_{\mathrm{\α}}\\ p{\hat{e}}_{\mathrm{\β}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}+{\mathrm{Ns}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, $s_{\mathrm{\α}}=M\mathrm{sgn}\left({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}}\right)+K\left({\hat{i}}_{\mathrm{\α}}-i_{\mathrm{\α}}\right),s_{\mathrm{\β}}=M\mathrm{sgn}\left({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}}\right)+K\left({\hat{i}}_{\mathrm{\β}}-i_{\mathrm{\β}}\right),$ M is sliding formwork gain, and sgn () is sign function, and K is state gain, and N is effective back electromotive force e _α, e _βthe gain of sliding mode observer, for the estimation current value on static alpha-beta coordinate system, for the effective back electromotive force estimated, for the synchronous speed estimated, L _qfor motor q axle inductance, R _sfor motor stator resistance;

(2) the effective back electromotive force estimated is obtained step as follows:

Step 1, if motor stator winding is delta connection, then performs step 1.1, if motor stator winding is star connection, then performs step 1.2,

Step 1.1, first sample motor stator voltage u _aB, u _bC, u _cA, then by u _aB, u _bC, u _cAcoordinate transform on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{u}_{\mathrm{\α}}\\ u_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{u}_{AB}\\ u_{BC}\\ u_{CA}\end{array}\right],$

Wherein, u _α, u _βfor the component of stator voltage on static alpha-beta coordinate system, afterwards, first sample motor stator current i _a, i _b, then calculate phase current, as shown in the formula:

$\left\{\begin{array}{c}{i}_{Ax}=1/3\left(i_A-i_B\right)\\ i_{Bx}=i_{Ax}+i_B\\ i_{Cx}=i_{Ax}-i_A\end{array}\right.,$

Wherein, i _ax, i _bx, i _cxfor three-phase phase current, finally, by the coordinate transform of three-phase phase current on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{i}_{\mathrm{\α}}\\ i_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{i}_{Ax}\\ i_{Bx}\\ i_{Cx}\end{array}\right],$

Wherein, i _α, i _βfor the component of phase current on static alpha-beta coordinate system;

Step 1.2, first sample motor stator voltage u _aB, u _bC, then calculate machine phase voltages, as shown in the formula:

$\left\{\begin{array}{c}{u}_B=1/3\left(u_{BC}-u_{AB}\right)\\ u_A=u_B+u_{AB}\\ u_C=u_B-u_{BC}\end{array}\right.,$

Wherein, u _a, u _b, u _cfor three-phase phase voltage, afterwards, by three-phase phase voltage coordinate transform on static alpha-beta coordinate system, as shown in the formula:

$\left[\begin{array}{c}{u}_{\mathrm{\α}}\\ u_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{u}_A\\ u_B\\ u_C\end{array}\right],$

Wherein, u _α, u _βfor the component of stator voltage on static alpha-beta coordinate system, finally, first sample motor stator current i _a, i _b, i _c, then obtain the component of electric current on static alpha-beta coordinate system through coordinate transform, as shown in the formula:

$\left[\begin{array}{c}{i}_{\mathrm{\α}}\\ i_{\mathrm{\β}}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \frac{\sqrt{3}}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{c}{i}_A\\ i_B\\ i_C\end{array}\right],$

Wherein, i _α, i _βfor the component of phase current on static alpha-beta coordinate system;

Step 2, first with the electric current estimated deduct the actual current i that step 1 obtains _α, i _β, obtain current deviation again respectively through sign function sgn () computing and multiplying, finally obtain current error feedback quantity s _α, s _β, as shown in the formula:

$s_{\mathrm{\α}}=M\mathrm{sgn}\left(\stackrel{\‾}{i_{\mathrm{\α}}}\right)+K\left(\stackrel{\‾}{i_{\mathrm{\α}}}\right),s_{\mathrm{\β}}=M\mathrm{sgn}\left(\stackrel{\‾}{i_{\mathrm{\β}}}\right)+K\left(\stackrel{\‾}{i_{\mathrm{\β}}}\right),$

Wherein, M is sliding formwork gain, and K is state gain,

Step 3, the voltage u obtained by step 1 _α, u _βdeduct the current error feedback quantity s that step 2 obtains _α, s _β, then deduct effective back electromotive force of the estimation that kth time computing obtains obtain first group of intermediate variable E _{1 α}, E _{1 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{1\mathrm{\α}}=u_{\mathrm{\α}}-s_{\mathrm{\α}}-{\hat{e}}_{\mathrm{\α}}\left(k\right)\\ E_{1\mathrm{\β}}=u_{\mathrm{\β}}-s_{\mathrm{\β}}-{\hat{e}}_{\mathrm{\β}}\left(k\right)\end{array}\right.,$

Wherein, E _{1 α}, E _{1 β}be first group of intermediate variable, k is positive integer, (k=0,1,2), ${\hat{e}}_{\mathrm{\β}}\left(0\right)=0;$

Step 4, the first group of intermediate variable E obtained by step 3 _{1 α}, E _{1 β}divided by motor q axle inductance, obtain second group of intermediate variable E _{2 α}, E _{2 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{2\mathrm{\α}}=E_{1\mathrm{\α}}/L_q\\ E_{2\mathrm{\β}}=E_{1\mathrm{\β}}/L_q\end{array}\right.,$

Wherein, L _qfor motor q axle inductance, E _{2 α}, E _{2 β}be second group of intermediate variable;

Step 5, according to the motor stator electric current estimated with motor stator resistance R _sand q axle inductance L _q, calculate the 3rd group of intermediate variable E _{3 α}, E _{3 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{3\mathrm{\α}}=\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\α}}\\ E_{3\mathrm{\β}}=\frac{R_s}{L_q}{\hat{i}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, E _{3 α}, E _{3 β}be the 3rd group of intermediate variable, R _sfor motor stator resistance;

Step 6, the E obtained by step 4 _{2 α}, E _{2 β}deduct the E that step 5 obtains _{3 α}, E _{3 β}, obtain the 4th group of intermediate variable E _{4 α}, E _{4 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{4\mathrm{\α}}=E_{2\mathrm{\α}}-E_{3\mathrm{\α}}\\ E_{4\mathrm{\β}}=E_{2\mathrm{\β}}-E_{3\mathrm{\β}}\end{array}\right.,$

Wherein, E _{4 α}, E _{4 β}be the 4th group of intermediate variable;

Step 7, to the E that step 6 obtains _{4 α}, E _{4 β}carry out integration, obtain the electric current estimated as shown in the formula:

$\left\{\begin{array}{c}{\hat{i}}_{\mathrm{\α}}={\hat{i}}_{\mathrm{\α}}+T_s{E}_{4\mathrm{\α}}\\ {\hat{i}}_{\mathrm{\β}}={\hat{i}}_{\mathrm{\β}}+T_s{E}_{4\mathrm{\β}}\end{array}\right.,$

Wherein, T _sfor the sampling period;

Step 8, the current estimation value that step 7 is obtained bring step 2 and step 5 into, for estimating stator current further;

Step 9, the current error feedback quantity s that step 2 is obtained _α, s _βbe multiplied by gain N, obtain the 5th group of intermediate variable E _{5 α}, E _{5 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{5\mathrm{\α}}={\mathrm{Ns}}_{\mathrm{\α}}\\ E_{5\mathrm{\β}}={\mathrm{Ns}}_{\mathrm{\β}}\end{array}\right.,$

Wherein, E _{5 α}, E _{5 β}be the 5th group of intermediate variable, N is effective back electromotive force e _α, e _βthe gain of sliding mode observer;

Step 10, the rotating speed of estimation be multiplied by effective back electromotive force of the estimation that kth time computing obtains respectively obtain the 6th group of intermediate variable E _{6 α}, E _{6 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{6\mathrm{\α}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\α}}\left(k\right)\\ E_{6\mathrm{\β}}={\widehat{\mathrm{\ω}}}_r{\hat{e}}_{\mathrm{\β}}\left(k\right)\end{array}\right.,$

Wherein, E _{6 α}, E _{6 β}be the 6th group of intermediate variable;

Step 11, the E first obtained by step 9 _{5 α}deduct the E that step 10 obtains _{6 β}, obtain the 7th intermediate variable E _{7 α}, then the E obtained by step 9 _{5 β}add the E that step 10 obtains _{6 α}, obtain the 7th intermediate variable E _{7 β}, as shown in the formula:

$\left\{\begin{array}{c}{E}_{7\mathrm{\α}}=E_{5\mathrm{\α}}-E_{6\mathrm{\β}}\\ E_{7\mathrm{\β}}=E_{5\mathrm{\β}}+E_{6\mathrm{\α}}\end{array}\right.$

Wherein, E _{7 α}, E _{7 β}be the 7th group of intermediate variable;

Step 12, respectively to the E that step 11 obtains _{7 α}, E _{7 β}carry out integration, obtain effective back electromotive force that kth+1 computing is estimated as shown in the formula:

$\left\{\begin{array}{c}{\hat{e}}_{\mathrm{\α}}\left(k+1\right)={\hat{e}}_{\mathrm{\α}}\left(k\right)+T_s{E}_{7\mathrm{\α}}\\ {\hat{e}}_{\mathrm{\β}}\left(k+1\right)={\hat{e}}_{\mathrm{\β}}\left(k\right)+T_s{E}_{7\mathrm{\β}}\end{array}\right.;$

Step 13, effective back electromotive force of the estimation first step 12 obtained bring step 3 into, be further used for solving first group of intermediate variable E _{1 α}, E _{1 β}, more effective back electromotive force of the estimation that step 12 is obtained bring step 10 into, for calculating the 7th group of intermediate variable E _{7 α}, E _{7 β};

Step 14, effective back electromotive force of the estimation that step 12 is obtained bring phase-locked loop into, obtain the rotor-position estimated with the rotating speed estimated

Step 15, by the rotating speed estimated bring step 10 into, for calculating the 7th group of intermediate variable E _{7 α}, E _{7 β};

Step 16, repeats step 1 ~ step 15, until the current value estimated equal actual current value i _α, i _β, obtain the back electromotive force e with reality _α, e _βeffective back electromotive force of the estimation conformed to and the rotor-position of the estimation to conform to the rotor position of reality

2. a kind of method realizing salient pole permanent magnet synchronous motor senseless control according to claim 1, is characterized in that phase-locked loop is according to the effective back electromotive force estimated obtain rotor-position with the rotating speed estimated step as follows:

Step 1, according to the effective back electromotive force of estimation calculate effective back electromotive force q axle deviation, as shown in the formula:

${\hat{e}}_q=-{\hat{e}}_{\mathrm{\α}}\left(k+1\right)\cos \widehat{\mathrm{\θ}}-{\hat{e}}_{\mathrm{\β}}\left(k+1\right)\sin \widehat{\mathrm{\θ}},$

Wherein, for effective back electromotive force q axle deviation;

Step 2, is obtained by step 1 the rotating speed estimated is obtained through proportional and integral controller as shown in the formula:

${\widehat{\mathrm{\ω}}}_r=\left(k_i/s+k_p\right){\hat{e}}_q,$

Wherein, k _pand k _ibe respectively ratio and integral coefficient, s is Laplacian;

Step 3, the estimation rotating speed obtained by step 2 the rotor-position estimated is obtained by integration as shown in the formula:

$\widehat{\mathrm{\θ}}=\widehat{\mathrm{\θ}}+T_s{\widehat{\mathrm{\ω}}}_r.$