ROBUST POSITION CONTROL METHOD FOR PERMANENT MAGNET SYNCHRONOUS MOTOR CONSIDERING CURRENT LIMITATION

Abstract

A robust position control method for a permanent magnet synchronous motor considering current limitation is provided. The method fully considers the influence of current limitation on a closed loop system in controller design, stability analysis and other theoretical analysis phase, can effectively overcome the influence of system disturbances including system parameters uncertainty and unknown load torque, and finally realize a control objective of accurate tracking of the motor position. More importantly, the technology is a continuous control method which can overcome the inherent chattering problem while having strong robustness of sliding mode control. Meanwhile, a controller designed by the present invention also has the advantages of simple structure, etc. The technical solution proposed by the present invention has wide practical application prospect due to the characteristics of excellent anti-disturbance capability, and simple and feasible structure.

Claims

1. A robust position control method for a permanent magnet synchronous motor (PMSM) considering current limitation, comprising steps of: step 1: determining a dynamic equation of PMSM influenced by amplitude limited element and disturbances: in an actual system, expressing a dynamic equation of a PMSM control system is as: $\stackrel{.Math.}{\theta }=\frac{K_t}{J}{i}_q-\frac{B}{J}\stackrel{.}{\theta }-\frac{T_L}{J};$ where θ represents motor rotor angle, i.sub.q represents q axis stator current in d-q coordinate system, K.sub.t represents torque constant, J represents moment of inertia of the motor, B represents viscous friction coefficient, T.sub.L represents load torque; further considering the influence of system parameters uncertainty, unknown load torque and current loop tracking error, rewriting the dynamic equation of the PMSM as: $\stackrel{.Math.}{\theta }=\frac{K_{to}}{J_o}{i}_q^{*}-\frac{B_o}{J_o}\stackrel{.}{\theta }+\left(\frac{K_{to}}{J_o}+\Delta \frac{K_t}{J}\right)\left(i_q-i_q^{*}\right)+\Delta \frac{K_t}{J}{i}_q^{*}-\Delta \frac{B}{J}\stackrel{.}{\theta }-\frac{T_L}{J};$ where i.sub.q* represents reference value of the stator current of q axis, K.sub.to, J.sub.O, and B.sub.o respectively represent nominal values of the torque constant, the moment of inertia, and the viscous friction coefficient, $\Delta \frac{K_t}{J}=\frac{K_t}{J}-\frac{K_{\mathrm{to}}}{J_o}$ and $\Delta \frac{B}{J}=\frac{B}{J}-\frac{B_o}{J_o}$ represent deviations between the true values of the system parameters and the nominal values; expressing the influence of the amplitude limited element on the reference current by the following equation: $i_q^{*}=f\left(u\right)=\{\begin{array}{cc}{I}_{\max }& u\ge I_{\max }\\ u& -I_{\max }<u<I_{\max }\\ -I_{\max }& u\le -I_{\max }\end{array};$ where u(t) represents control input to be designed, i.e. a PMSM position loop controller, I.sub.max represents limitation value of the amplitude limited element; thus, the following relation holds: i.sub.q*=f(u)=u+Δu; where Δu=f(u)−u represents influence caused by the amplitude limited element; obtaining a complete dynamic equation of the PMSM comprehensively considering influence of the system disturbances and amplitude limited element: $\stackrel{.Math.}{\theta }=\frac{K_{\mathrm{to}}}{J_o}u-\frac{B_o}{J_o}\stackrel{.}{\theta }-d;$ where d(t) represents lumped disturbances term of which the specific expression is: $d=-\left(\frac{K_{to}}{J_o}+\Delta \frac{K_t}{J}\right)\left(i_q-i_q^{*}\right)-\Delta \frac{K_t}{J}{i}_q^{*}+\Delta \frac{B}{J}\stackrel{.}{\theta }+\frac{T_L}{J}-\frac{K_{to}}{J_o}\Delta u;$ step 2: determining a control objective and constructing auxiliary signals: in PMSM position control, it is guaranteed that the motor rotor angle can accurately reach a given position within a limited time, i.e. $\underset{t.fwdarw.\infty }{\lim }\theta \left(t\right)={\theta }_d;$ where θ.sub.d(t) represents target rotor position of the PMSM; further defining position tracking error signal as e.sub.1=θ.sub.d−θ; for the follow-up controller design and stability analysis, constructing auxiliary signals of the following forms: e.sub.2=ė.sub.1+αe.sub.1, r=ė.sub.2+βe.sub.2; where both α and β are positive constants greater than 0; step 3: designing a robust position controller and conducting stability analysis of the closed loop system: based on step 1 and step 2, giving a PMSM robust position controller of the following form: $\begin{array}{cc}u=\frac{J_o}{K_{to}}\left[\left(k+1\right)e_2-\left(k+1\right)e_2\left(0\right)+{\int }_0^t\left[\left(k+1\right)\beta e_2\left(\tau \right)+2\lambda \mathrm{sgn}\left(e_2\left(\tau \right)\right)\right]d\tau \right];& \end{array}$ where k and λ represent positive control gains; constructing a Lyapunov function candidate: $V=\frac{1}{2}{r}^2+\frac{1}{2}{e}_1^2+\frac{1}{2}{e}_2^2+2\lambda .Math.e_2.Math.-{\mathrm{Ne}}_2;$ in combination with the Lyapunov stability method and LaSalle-Yoshizawa theorem, proving asymptotic stability of the closed loop system; step 4: measuring the position and speed of the motor in real time by a sensor installed in the PMSM first, obtaining system state variables and substituting same into the robust position controller given in step 3 to obtain control signal, taking the control signal as controller of a position loop of the PMSM, so as to accurately track the position of the motor rotor and effectively suppress the influence of disturbances including system parameters uncertainty, unknown load torque and current limitation to ensure that a PMSM system can still realize a quick and accurate positioning function under the influence of disturbances.

Description

DESCRIPTION OF DRAWINGS

[0033] FIG. 1 is a block diagram of robust position control of a PMSM proposed by the present invention;

[0034] FIG. 2 is a block diagram of an industrial common three closed loop position control of PMSM;

[0035] FIG. 3(a) shows a response curve of a rotor angle θ of a PMSM under the control of the method proposed by the present invention in an ideal condition;

[0036] FIG. 3(b) shows a curve of a q axis stator current i.sub.q of a PMSM in the method proposed by the present invention in an ideal condition;

[0037] FIG. 4(a) shows a response curve of a rotor angle θ of a PMSM under the control of an industrial common method in an ideal condition;

[0038] FIG. 4(b) shows a curve of a q axis stator current i.sub.q of a PMSM in an industrial common method in an ideal condition;

[0039] FIG. 5(a) shows a response curve of a rotor angle θ of a PMSM under the control of the method proposed by the present invention under disturbances influence;

[0040] FIG. 5(b) shows a curve of a q axis stator current i.sub.q of a PMSM in the method proposed by the present invention under disturbances influence;

[0041] FIG. 6(a) shows a response curve of a rotor angle θ of a PMSM under the control of an industrial common method under disturbances influence;

[0042] FIG. 6(b) shows a curve of a q axis stator current i.sub.q of a PMSM in an industrial common method under disturbances influence.

DETAILED DESCRIPTION

[0043] The technical solution proposed by the present invention is further described below in detail in combination with the drawings and specific embodiments.

Embodiment 1

[0044] As shown in FIG. 1, this embodiment discloses a robust position control method for PMSM considering current limitation, comprising the following steps:

(I) Determining a Dynamic Equation of a PMSM Influenced by Amplitude Limited Element and Disturbances:

[0045] an object studied in this technical solution is a surface-mounted PMSM and is based on the i.sub.d=0 vector control framework as shown in FIG. 1, this control framework takes the rotor coordinate system (d-q coordinate system) as a reference coordinate system, under this coordinate system, the voltage equation of the system is as follows:

u.sub.d=Li.sub.d+Ri.sub.d−Lnωi.sub.q

u.sub.q=Li.sub.q+Ri.sub.q+Lnωi.sub.d−nψ.sub.fω  (1)

where u.sub.d and u.sub.q represent components of stator voltages on d axis and q axis, i.sub.d, i.sub.q represent stator currents on d axis and q axis respectively, meanwhile, R and L represent stator resistance and stator inductance, n represents the number of the pole pairs of the motor, ψ.sub.f represents a permanent magnet flux linkage of the rotor, and ω represents a speed of the motor.

[0046] The expression of the electromagnetic torque of the surface-mounted PMSM is as follows:

T.sub.e=K.sub.ti.sub.q  (2)

where T.sub.e represents electromagnetic torque, and K.sub.t represents a torque constant.

[0047] Next, a motion equation of the PMSM system is given as follows:

[00011]$\begin{array}{cc}{T}_e-T_L=J\frac{d^2}{{\mathrm{dt}}^2}\theta +B\frac{d}{\mathrm{dt}}\theta & \left(3\right)\end{array}$

where T.sub.L represents load torque, θ represents a motor rotor angle, J represents a motor moment of inertia and B represents a viscous friction coefficient.

[0048] In combination with equation (2) and equation (3), a state equation of the PMSM control system can be obtained:

[00012]$\begin{array}{cc}\stackrel{.Math.}{\theta }=\frac{K_t}{J}{i}_q-\frac{B}{J}\stackrel{.}{\theta }-\frac{T_L}{J}& \left(4\right)\end{array}$

[0049] It should be pointed out that the parameters in the above equation are all actual system parameters and the true values of these parameters are often difficult to obtain in practical application, so researchers can only obtain the nominal values of relevant parameters. Therefore, by further considering the influence of system parameters uncertainty, unknown load torque and current loop tracking error, equation (4) can be rewritten as:

[00013]$\begin{array}{cc}\stackrel{.Math.}{\theta }=\frac{K_{to}}{J_o}{i}_q^{*}-\frac{B_o}{J_o}\stackrel{.}{\theta }+\left(\frac{K_{to}}{J_o}+\Delta \frac{K_t}{J}\right)\left(i_q-i_q^{*}\right)+\Delta \frac{K_t}{J}{i}_q^{*}-\Delta \frac{B}{J}\stackrel{.}{\theta }-\frac{T_L}{J}& \left(5\right)\end{array}$

where i.sub.q* represents a reference value of the stator current of q axis, K.sub.to, J.sub.o, and B.sub.o respectively represent nominal values of the torque constant, the moment of inertia, and the viscous friction coefficient,

[00014]$\Delta \frac{K_t}{J}=\frac{K_t}{J}-\frac{K_{to}}{J_o}$

and

[00015]$\Delta \frac{B}{J}=\frac{B}{J}-\frac{B_o}{J_o}$

represent deviations between the true values of the system parameters and the nominal values.

[0050] It can be known in combination with FIG. 1 that in order to prevent the motor current from exceeding the safety limit, a control input u generated by a position loop controller cannot be directly sent to the current inner loop as the reference value of the stator current of q axis, but an amplitude limited element needs to be applied after the position loop controller u to ensure that the given value of the motor current is constrained within a given range, thus realizing indirect limitation of the motor current. It is easy to know that when exceeding the limit of the amplitude limited element, the output value of the position loop controller may be limited, which may result in a deviation between the output value of the position loop and the current reference value actually sent to the current inner loop. Unlike most of the existing technical solutions, the present invention may consider the influence of the amplitude limited element on the closed loop system rather than simply neglecting the amplitude limited element.

[0051] The influence of the amplitude limited element on the reference current can be expressed by the following equation:

[00016]$\begin{array}{cc}{i}_q^{*}=f\left(u\right)=\{\begin{array}{cc}{I}_{\max }& u\ge I_{\max }\\ u& -I_{\max }<u<I_{\max }\\ -I_{\max }& u\le -I_{\max }\end{array}& \left(6\right)\end{array}$

where u(t) represents a control input to be designed, i.e. a PMSM position loop controller, I.sub.max represents a limitation value of the amplitude limited element. Then, the following relation holds:

i.sub.q*=f(u)=u+Δu  (7)

where Δu=f(u)−u represents the influence caused by the amplitude limited element.

[0052] In combination with equations (5), (6) and (7), a complete dynamic equation of the PMSM which comprehensively considers the influence of the internal and external system disturbances and the amplitude limited element can be obtained:

[00017]$\begin{array}{cc}\stackrel{.Math.}{\theta }=\frac{K_{to}}{Jo}u-\frac{B_o}{J_o}\stackrel{.}{\theta }-d& \left(8\right)\end{array}$

where d(t) represents a lumped disturbances term of which a specific expression t is

[00018]$\begin{array}{cc}d=-\left(\frac{K_{to}}{J_o}+\Delta \frac{K_t}{J}\right)\left(i_q-i_q^{*}\right)-\Delta \frac{K_t}{J}{i}_q^{*}+\Delta \frac{B}{J}\stackrel{.}{\theta }+\frac{T_L}{J}-\frac{K_{to}}{J_o}\Delta u& \left(9\right)\end{array}$

for lumped disturbances d(t) and first and second derivatives thereof, the following bounded assumption is usually made:

d,{dot over (d)},{umlaut over (d)}∈ζ.sub.∞  (10)

(II): Determining a Control Objective and Constructing Auxiliary Signals:

[0053] It is assumed that θ.sub.d(t) represents a target rotor position of the PMSM, that is, a position given signal, and it is assumed that a continuous third derivative thereof is bounded, that is,

θ.sub.d,{dot over (θ)}.sub.d,{umlaut over (θ)}.sub.d,.sub.d∈ζ.sub.∞  (11)

[0054] in the position control of the PMSM, the main object is to ensure that the motor rotor angle can reach the given position accurately in a limited time, that is

[00019]$\begin{array}{cc}\underset{t.fwdarw.\infty }{\lim }\theta \left(t\right)={\theta }_d& \left(12\right)\end{array}$

[0055] The position tracking error signal can be further defined as

e.sub.1=θ.sub.d−θ  (13)

[0056] On this basis, in order to facilitate follow-up controller design and stability analysis, auxiliary signals of the following forms are constructed

e.sub.2=ė.sub.1+αe.sub.1,r=ė.sub.2+βe.sub.2  (14)

where both α and β are positive constants greater than 0.

[0057] According to equations (8), (13) and (14), it can be obtained that:

[00020]$\begin{array}{cc}r=-\frac{K_{to}}{J_o}u+{\stackrel{.Math.}{\theta }}_d+\frac{B_o}{J_o}{\stackrel{.}{\theta }}_d+d+\left(\alpha +\frac{B_o}{J_o}\right){\stackrel{.}{e}}_1+\beta e_2& \left(15\right)\end{array}$

by taking the derivative of the above equation and transforming same, it is easy to know that:

[00021]$\begin{array}{cc}\stackrel{.}{r}=-\frac{K_{to}}{J_o}\stackrel{.}{u}-e_2-r+\left(\alpha +\frac{B_o}{J_o}\right){\stackrel{.Math.}{e}}_1+\beta {\stackrel{.}{e}}_2+e_2+r+{\stackrel{}{\theta }}_d+\frac{B_o}{J_o}{\stackrel{.Math.}{\theta }}_d+\stackrel{.}{d}& \left(16\right)\end{array}$

if

[00022]$\begin{array}{cc}H=\left(\alpha +\frac{B_o}{J_o}\right){\stackrel{.Math.}{e}}_1+\beta {\stackrel{.}{e}}_2+e_2+r_2& \left(17\right)\end{array}$$\begin{array}{cc}N={\stackrel{}{\theta }}_d+\frac{B_o}{J_o}{\stackrel{.Math.}{\theta }}_d+\stackrel{.}{d}& \left(18\right)\end{array}$$\mathrm{then}$$\begin{array}{cc}\dot{r}=-\frac{K_{to}}{J_o}\stackrel{.}{u}-e_2-r+H+N& \left(19\right)\end{array}$

[0058] The boundednesses of H(t) and N(t) are analyzed below. First, the boundedness of N(t) is analyzed, according to equations (10) and (11), it is easy to obtain:

∥N∥.sub.∞≤ε.sub.1,∥{dot over (N)}∥.sub.∞≤ε.sub.2  (20)

where ε.sub.1 and ε.sub.1 are positive constants.

[0059] Then, the boundedness of H(t) is analyzed, and according to equation (14), if

ė.sub.2=r−βe.sub.2,ë.sub.1=ė.sub.2−αė.sub.1=r−βe.sub.2−α(e.sub.2−αe.sub.1)=r−(α+β)e.sub.2+α.sup.2e.sub.1  (21)

H(t) can be rewritten as

H(t)=q.Math.Z  (22)

where Z=(r, e.sub.1, e.sub.2).sup.T and

[00023]$\begin{array}{cc}q={\left[\begin{array}{c}\left(\alpha +\frac{B_o}{J_o}\right)+\beta +1\\ \left(\alpha +\frac{B_o}{J_o}\right){\alpha }^2\\ -\left(\alpha +\frac{B_o}{J_o}\right)\left(\alpha +\beta \right)-{\beta }^2+1\end{array}\right]}^T& \left(23\right)\end{array}$

it can be known that

∥H∥.sub.∞=|H|=∥q∥∥Z∥≤ρ∥Z∥  (24)

where ρ≥∥q∥ is a positive constant.

(III) Designing a Robust Position Controller and Conducting Stability Analysis of the Closed Loop System:

[0060] on the basis of the first two parts, the present invention provides a robust position controller for a PMSM as follows:

[00024]$\begin{array}{cc}u=\frac{J_o}{K_{to}}\left[\text{⁠}\left(k+1\right)e_2-\left(k+1\right)e_2\left(0\right)+{\int }_0^t\left[\left(k+1\right)\beta e_2\left(\tau \right)+2\mathrm{\lambda sgn}\left(e_2\left(\tau \right)\right)\right]d\tau \right]& \left(25\right)\end{array}$

where k and λ are positive adjustable control gains.

[0061] The controller in (25) is substituted into equation (19), obtaining

{dot over (r)}=−(k+1)r−2λsgn(e.sub.2)−e.sub.2−r+N+H  (26)

[0062] System stability analysis is conducted below to prove that the system control object mentioned in part (II) can be achieved. First, the following theorem is given:

Theorem: when the following condition

[00025]$\begin{array}{cc}k\ge \frac{{\rho }^2}{l}\left(l=\min \left(2,\left(\alpha -\frac{1}{2}\right),\left(\beta -\frac{1}{2}\right)\right)\right),\lambda >{\epsilon }_1+\frac{{\epsilon }_2}{\beta },\alpha >\frac{1}{2},\beta >\frac{1}{2}& \left(27\right)\end{array}$

holds, the rotor angle of the PMSM will accurately reach a target position under the action of the controller in (25) designed by the present invention, that is,

[00026]$\begin{array}{cc}\underset{t.fwdarw.\infty }{\lim }\theta \left(t\right)={\theta }_d& \left(28\right)\end{array}$

Proof: a Lyapunov candidate function of the following form is constructed:

[00027]$\begin{array}{cc}V=\frac{1}{2}{r}^2+\frac{1}{2}{e}_1^2+\frac{1}{2}{e}_2^2+2\lambda .Math.e_2.Math.-{\mathrm{Ne}}_2& \left(29\right)\end{array}$

Definition

[0063]
Λ=2λ|e.sub.2|−Ne.sub.2  (30)

The following analysis shows that it is always greater than 0. It can be known from Ne.sub.2≤∥N∥.sub.∞|e.sub.2| that

−Ne.sub.2≥−∥N∥.sub.∞|e.sub.2|  (31)

In combination with equation (20) and gain condition in (27), it can be deduced that

2λ|e.sub.2|−∥N∥.sub.∞|e.sub.2|≥λ|e.sub.2|≥0  (32)

Then

Λ=2λ|e.sub.2|−Ne.sub.2≥2λ|e.sub.2|−∥N∥.sub.∞|e.sub.2|≥λ|e.sub.2|≥0  (33)

Meanwhile, if −Ne.sub.2≤∥N∥.sub.∞|e.sub.2|, then

Λ=2λ|e.sub.2|−Ne.sub.2≤(2λ+∥N∥.sub.∞)|e.sub.2|  (34)

In combination with equations (33) and (34), it is obtained that

0≤λ|e.sub.2|≤2λ|e.sub.2|−∥N∥.sub.∞|e.sub.2|≤Λ×(2λ+∥N∥.sub.∞)|e.sub.2|  (35)

Further, according to the form of equation (29), it can be known that

[00028]$\begin{array}{cc}\frac{1}{2}{Z}^2+\lambda .Math.e_2.Math.\le V\le \frac{1}{2}{Z}^2+\left(2\lambda +{.Math.N.Math.}_{\infty }\right).Math.e_2.Math.& \left(36\right)\end{array}$

The above result shows that the designed V(t) is non-negative, so it can be used as a Lyapunov function. By taking the derivative of the Lyapunov function, in combination with equations (14) and (26), it can be obtained that

[00029]$\begin{array}{cc}\begin{array}{c}\stackrel{.}{V}=r\dot{r}+e_1{\stackrel{.}{e}}_1+e_2{\stackrel{.}{e}}_2+2\lambda {\stackrel{.}{e}}_2\mathrm{sgn}\left(e_2\right)-\stackrel{.}{N}{e}_2-N{\stackrel{.}{e}}_2\\ =-\left(k+1\right)r^2-2\lambda r\mathrm{sgn}\left(e_2\right)-e_{2}r-r^2+\mathrm{rN}+\\ \mathrm{rH}+e_1\left(e_2-\alpha e_1\right)+e_2\left(r-\beta e_2\right)+2\lambda {\stackrel{.}{e}}_2\mathrm{sgn}\left(e_2\right)-\stackrel{.}{N}{e}_2-N{\stackrel{.}{e}}_2\\ =-\left(k+1\right)r^2-2\lambda {\stackrel{.}{e}}_2\mathrm{sgn}\left(e_2\right)-2\lambda \beta e_2\mathrm{sgn}\left(e_2\right)-\\ e_{2}r-r^2+N{\stackrel{.}{e}}_2+\beta N{e}_2+rH-\alpha e_1^2+e_1{e}_2-\beta e_2^2+\\ e_{2}r+2\lambda {\stackrel{.}{e}}_2\mathrm{sgn}\left(e_2\right)-\stackrel{.}{N}{e}_2-N{\stackrel{.}{e}}_2\\ =-\left(k+1\right)r^2-2\lambda \beta e_2\mathrm{sgn}\left(e_2\right)-\\ r^2+\beta N{e}_2+rH-\alpha e_1^2+e_1{e}_2-\beta e_2^2-\stackrel{.}{N}{e}_2\end{array}& \left(37\right)\end{array}$

because

[00030]$\begin{array}{cc}{e}_2\mathrm{sgn}\left(e_2\right)=.Math.e_2.Math.,e_1{e}_2\le \frac{1}{2}{e}_1^2+\frac{1}{2}{e}_2^2,\beta {\mathrm{Ne}}_2\le \beta {.Math.N.Math.}_{\infty }.Math.e_2.Math.,\stackrel{.}{N}{e}_2\le {.Math.\stackrel{.}{N}.Math.}_{\infty }.Math.e_2.Math.,\mathrm{rH}\le .Math.r.Math.{.Math.H.Math.}_{\infty }& \left(38\right)\end{array}$

equation (37) can be simplified as

[00031]$\begin{array}{cc}\stackrel{.}{V}\le -\left(k+1\right)r^2-2\mathrm{\lambda \beta }.Math.e_2.Math.-r^2+\beta {.Math.N.Math.}_{\infty }.Math.e_2.Math.+.Math.r.Math.{.Math.H.Math.}_{\infty }-\left(\alpha -\frac{1}{2}\right)e_1^2-\left(\beta -\frac{1}{2}\right)e_2^2-{.Math.\stackrel{.}{N}.Math.}_{\infty }.Math.e_2.Math.& \left(39\right)\end{array}$

Further, by equations (20), (24) and (27), it can be obtained that

−λβ|e.sub.2|≤−(βε.sub.1+ε.sub.2)|e.sub.2|,∥{dot over (N)}∥.sub.∞|e.sub.2|≤ε.sub.2|e.sub.2|,β∥N∥.sub.∞|e.sub.2|≤βε.sub.1|e.sub.2|,|r|∥H∥.sub.∞≤ρ∥Z∥|r|  (40)

Thus, equation (39) can be rewritten as

[00032]$\begin{array}{cc}\stackrel{.}{V}\le \left[-2r^2-\left(\alpha -\frac{1}{2}\right)e_1^2-\left(\beta -\frac{1}{2}\right)e_2^2\right]+\rho .Math.Z.Math..Math.r.Math.-{\mathrm{kr}}^2-\mathrm{\lambda \beta }.Math.e_2.Math.\le \left[-2r^2-\left(\alpha -\frac{1}{2}\right)e_1^2-\left(\beta -\frac{1}{2}\right)e_2^2\right]+\frac{{\rho }^2{.Math.Z.Math.}^2}{k}-\frac{{\rho }^2{.Math.Z.Math.}^2}{k}+{\rho }^2{.Math.Z.Math.}^2.Math.r.Math.-{\mathrm{kr}}^2-\mathrm{\lambda \beta }.Math.e_2.Math.\le -\left(l-\frac{{\rho }^2}{k}\right){.Math.z.Math.}^2& \left(41\right)\end{array}$

where

[00033]$l=\min \left(2,\left(\alpha -\frac{1}{2}\right),\left(\beta -\frac{1}{2}\right)\right),$

if

[00034]$k\ge \frac{{\rho }^2}{l},$

the following equation holds

{dot over (V)}≤0  (42)

According to the results of equations (36) and (42), it can be obtained that

V,r,e.sub.1,e.sub.2εζ.sub.∞  (43)

According to the expression of the controller in (25), it can be further known that

u∈ζ.sub.∞  (44)

The above results show that both the signals and the control input in the closed loop system are bounded. Next, according to equations (29) and (42), by using LaSalle Yoshizawa Theorem, it can be obtained that

[00035]$\begin{array}{cc}\underset{t.fwdarw.\infty }{\lim }.Math.Z.Math.=0.Math.\underset{t.fwdarw.\infty }{\lim }{e}_1=0& \left(45\right)\end{array}$

that is,

[00036]$\begin{array}{cc}\underset{t.fwdarw.\infty }{\lim }\theta \left(t\right)={\theta }_d& \left(46\right)\end{array}$

The theorem is proved, that is, the rotor angle of the PMSM is accurately traced to a given target position.

(IV) Realizing a Technical Solution:

[0064] Here, the present invention briefly introduces how to use the method in actual industry. First, the position and speed of the motor are measured in real time by a sensor installed in the PMSM first, system state variables are obtained and substituted into the robust position controller in (25) designed by the present invention to obtain a control signal, the control signal is taken as a controller of a position loop of the PMSM shown in FIG. 1, so as to accurately track the position of the motor rotor and effectively suppress the influence of disturbances including system parameters uncertainty, unknown load torque and current limitation to ensure that a PMSM system can still realize a quick and accurate positioning function under the influence of disturbances.

[0065] Simulation verification: FIG. 1 is a structural diagram of the technology proposed by the present invention. The designed robust controller is used in the position loop and the classical PI controller is used in the current loop. FIG. 2 shows one of the most common control frameworks in industrial application. A P controller is used in the position loop, and PI controllers are used in both the speed loop and current loop. In the present invention, the two control schemes are simulated and compared to verify the effectiveness and the superiority of the technology proposed by the present invention.

Simulation 1: Position Tracking Performance of the Technology Proposed in an Ideal Condition

[0066] The control performance of the proposed method in an ideal condition is considered in this simulation, that is, the true values of the system parameters are known and equal to the nominal values thereof, and no influence of external disturbances such as load torque change, etc. are present. In this simulation, system parameters are set as: J=J.sub.o=0.011 kg.Math.m.sup.2, B=B.sub.o=0.005N.Math.m.Math.s/rad, K=K.sub.to=3.6 N.Math.m/A, the load torque T.sub.L=4.5 N.Math.m, and the limitation value of the current amplitude limited element is ±10 A. The simulation results are shown in FIGS. 3-4, wherein FIG. 3(a) and FIG. 3(b) show simulation results of the method proposed by the present invention, FIG. 4(a) and FIG. 4(b) show simulation results of the industrial common scheme, the solid lines in FIG. 3(a) and FIG. 4(a) represent response curves of the rotor of the PMSM and the dashed line represents the target position θ.sub.d=3π of the rotor, and FIG. 3(b) and FIG. 4(b) respectively show the curves of the q axis stator current i.sub.q of the method proposed by the present invention and the industrial common method. It can be seen from FIG. 3(b) and FIG. 4(b) that the amplitude limited element has a constraint effect, so i.sub.q is limited within ±10 A, that is, the influence of the amplitude limited element is considered in this simulation. Further, it can be known by comparing FIG. 3(a) and FIG. 4 (a) that the method proposed by the present invention makes the motor rotor angle accurately reach the given target position at about 0.45 s, while the reaching time of the industrial common scheme is above 0.65 s, indicating that compared with the industrial common scheme, the PMSM position controller designed by the present invention has faster adjustment speed, so that the PMSM system can obtain better dynamic performance.

Simulation 2: Position Tracking Performance of the Technology Proposed Under Disturbance Influence

[0067] Further, in order to verify the robustness of the proposed method, the influence of internal and external disturbances including system parameters uncertainty, sudden change of outside load torque and the like is considered in simulation 2, the moment of inertia and the viscous friction coefficient are adjusted to J=0.022 kg.Math.m.sup.2, B=0.025N.Math.m.Math.s/rad, and other system parameters and controller parameters are unchanged. Meanwhile, in order to simulate the load torque change phenomenon, adjusting the load torque to 9N.Math.m at 0.8 s and adjusting the load torque back to 4.5N.Math.m at 1 s are simulated. The simulation results are shown in FIGS. 5-6, wherein FIG. 5(a) and FIG. 5(b) show simulation results of the method proposed by the present invention, and FIG. 6(a) and FIG. 6(b) show simulation results of the industrial common scheme. Similarly, the solid lines in FIG. 5(a) and FIG. 6(a) represent response curves of the rotor of the PMSM, while the dotted line represents the target position θ.sub.d=3π of the rotor, and FIG. 5(b) and FIG. 6(b) respectively show curves of q axis stator currents i.sub.q of the method proposed by the present invention and the industrial common method. FIG. 5(b) and FIG. 6(b) show that the role of the amplitude limited element is still taken into account in simulation 2. It can be known from FIG. 5(a) that when the influence of the amplitude limited element and the change of the system parameters are present, the method proposed by the present invention can still maintain good position tracking performance, and the motor rotor still reaches the given position at about 0.45 s. Meanwhile, when the load torque suddenly changes, under the adjustment action of the controller designed by the present invention, the motor rotor fluctuates slightly in position, the fluctuation range being only ±0.1 rad, and soon returns to the stable state, the position tracking error re-converges to 0. It can be seen from the dynamic response curve of the rotor of the industrial common scheme in FIG. 6(a), when system parameters change, the control effect of the industrial common three closed loop control scheme is reduced sharply, and the motor rotor angle is overshot, and is only stabilized at the given target position at about 0.7 s. Meanwhile, when the load torque suddenly changes, the motor rotor obviously fluctuates in position, the fluctuation range reaching ±0.7 rad which is far greater than the constraint range of the method proposed by the present invention.

[0068] In summary, the results of simulation 1 and simulation 2 show that compared with the three closed loop control scheme commonly used in industry, the technical solution proposed by the present invention has a faster dynamic response and can realize position tracking control of the PMSM within a short time, more importantly, the proposed technical solution has strong robustness to system parameters uncertainty, unknown load torque disturbance, etc., and can still ensure good position control performance under the change of system parameters and load torque, which means that the present invention has an important practical application prospect and can be applied to actual industrial production.