KR20160032894A - Torque control mutual indectance of interior permanent magnet synchronous motor - Google Patents

Abstract

The present invention relates to an apparatus and a method for controlling torque using mutual inductance estimation. More specifically, an apparatus for controlling torque using mutual inductance estimation to control the torque of an interior permanent magnet synchronous motor (IPMSM) comprises: a mutual inductance estimation unit which estimates a d-axis mutual inductance value and a q-axis mutual inductance value in accordance with an arbitrary torque instruction; a torque value estimation unit which estimates the torque control value on the basis of the mutual inductance values estimated in the mutual inductance estimation unit; a d-q axis current instruction value generation unit which generates a d-axis current instruction value and a q-axis current instruction value on the basis of the torque control value estimated in the torque value estimation unit; and a current controller which controls torque of the IPMSM by applying a current value to the IPMSM on the basis of the current instruction value generated in the d-q axis current instruction value.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a torque control device and a torque control method using mutual inductance estimation,

The present invention relates to a torque control apparatus and method using mutual inductance estimation.

Recently, due to the increase in interest in international oil prices and environmental pollution, the emission standards and fuel consumption standards required for passenger cars are increasing. It is attracting attention as an eco-friendly future vehicle that can satisfy the reduction of harmful emissions due to the introduction of hybrid cars and electric vehicles.

The hybrid vehicle can improve the operation efficiency of the automobile by complementary operation between the engine and the electric motor of the internal combustion engine. This can be accomplished by compensating the engine power by using the electric motor in the low speed region where the efficiency of the engine is low, So that the fuel efficiency is improved. In addition, electric vehicles are attracting attention as eco-friendly cars that do not have harmful emissions even though their mileage is small.

These eco-friendly automotive motors are mainly applied to permanent magnet permanent magnet synchronous motors (IPMSM) which have high efficiency and can operate at wide variable speed. Since the permanent magnet synchronous motor has no excitation winding for generating the magnetic flux in the rotor, there is no loss due to this, and the efficiency is good. The high output density has a large ratio of the output torque to the motor weight and the quick response is good. This characteristic is used in a high-performance motor control field requiring instantaneous torque control.

Therefore, it is essential to control the torque of the permanent magnet type permanent magnet motor used in the environment friendly automobile. In order to generate the maximum torque per unit current used in the torque control, the ratio of the magnetic torque due to the permanent magnet and the reluctance torque due to the difference in reluctance should be appropriately adjusted according to the load.

However, the value of the inductance influencing the torque is decreased by saturation of the stator and the rotor iron core while the temperature of the motor and the applied current increase. Therefore, the maximum torque operating point per unit current is changed according to the change of the current applied to the motor.

In addition, DTC (Direct Torque Control) is used for existing torque control. DTC is used in compact system because it has excellent transient response characteristic and simple principle because it uses switching table.

However, the switching frequency is not constant, and the switching frequency changes depending on the back electromotive force or the load variation. In addition, there is a disadvantage that a limit cycle section in which the switching frequency increases sharply occurs in the low-speed operation region.

As a result, there is a disadvantage in using the DTC, and torque control using the controller is required. The current command of the current controller uses the reference table method to ensure that the torque command and actual torque are accurate in the entire operating range. Through the experiment, the reference table method generates the maximum torque operating point according to the d-q axis current as a reference table over the entire operation range of the motor.

Using the reference table created from the experiment, the current command corresponding to the torque command according to the speed is used as the output. However, the reference table method requires a load testing device having a wider operating range than the driving motor, and has a disadvantage in that it performs many experiments.

Korean Patent No. 10-1353583 Korean Patent No. 10-1046042 Korean Patent No. 10-1221216 Korean Patent No. 10-0795283

SUMMARY OF THE INVENTION Accordingly, the present invention has been made keeping in mind the above problems occurring in the prior art, and it is an object of the present invention to provide a permanent magnet synchronous motor for an environmentally friendly automobile, which estimates mutual interference inductance of the eco- The present invention provides a control method capable of torque control.

Other objects, specific advantages and novel features of the present invention will become more apparent from the following detailed description and preferred embodiments with reference to the accompanying drawings.

A mutual inductance estimating unit for estimating a d-axis mutual inductance value and a q-axis mutual inductance value according to a torque command value, comprising: a mutual inductance estimating unit for estimating a d-axis mutual inductance value and a q-axis mutual inductance value according to a torque command value; A torque value estimator for estimating a torque control value based on the mutual inductance estimated value estimated by the mutual inductance estimating unit; A d-q-axis current command value generating unit for generating a d-axis current command value and a q-axis current command value based on the estimated torque control value estimated by the torque value estimating unit; And a current controller for controlling the torque of the IPMSM by applying a current value to the IPMSM based on the current command value generated by the dq axis current command value generator. . ≪ / RTI >

The current command generator may generate a current command value according to Equation (1) by applying a maximum torque per unit current (MTPA) control technique considering mutual inductance.

[Equation 1]

The mutual inductance estimation value is determined by the following equation (2).

&Quot; (2) "

Further, the torque estimation value may be determined by the following equation (3).

&Quot; (3) "

Further, the current controller is at least one of an anti-windup PI controller and a cascaded PI controller considering mutual interference of IPMSMs.

According to an embodiment of the present invention, the mutual interference inductance of the permanent magnet synchronous motor for a green vehicle is estimated and the torque can be calculated and the torque can be controlled based on the torque.

Also, according to the embodiment of the present invention, the mutual interference inductance is estimated, and the torque control value is estimated and controlled. Therefore, it is not necessary to generate the reference table through the existing experiment and make the command value of the current to fall within the error range There is an advantage that a motor having a larger capacity than the drive motor for carrying out the experiment does not need to be used as a load motor.

Although the present invention has been described in connection with the above-mentioned preferred embodiments, it will be appreciated by those skilled in the art that various other modifications and variations can be made without departing from the spirit and scope of the invention, All fall within the scope of the appended claims.

FIG. 1 is a graph showing FEM analysis results for dq axis magnetic flux variation,
FIG. 2 is a graph showing the dq axis magnetic flux and torque value according to the q-axis current,
3 is a graph of a d-axis magnetic flux value according to dq axis current,
4 is a graph showing a q-axis magnetic flux value according to the dq axis current,
5 is a graph showing the meaning of the q-axis magnetic flux parameter,
6 is a graph showing the meaning of the d-axis magnetic flux parameter,
7 is a block diagram of a typical current PI controller,
8 is a block diagram of a conventioanl PT controller,
9 is a Simulink block diagram of a conventioanl PT controller,
10 is a block diagram of a conventioanl PT controller without consideration of interference components,
11 is a Bode plot of a conventioanl PT controller that does not consider an interference component,
12 is a block diagram of a conventioanl PT controller considering interference components,
13 shows the Bode plot of the conventioanl PT controller considering the interference component.
Figure 14 is a block diagram of a Cascade PI controller,
Figure 15 shows the Bode plot of a cascaded PI controller,
16 shows the Bode plot of the cascaded PI controller at a control error of +/- 20%
17 is a graph showing changes in torque, voltage, current change,
Foo,
18 is a graph showing a stator current curve and a torque curve for representing the maximum torque per unit current at point A where the stator current and the torque curve are in contact with each other,
FIG. 19 is a graph showing the voltage limiting ellipse and the current limiting circle graph,
20 is an overall block diagram using a voltage regulator,
21 is a graph showing the dq axis current for maximum torque control per unit current,
22 is a block diagram of a back electromotive force estimation circuit,
23 shows a torque control block diagram.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the detailed description of known functions and configurations incorporated herein will be omitted when it may unnecessarily obscure the subject matter of the present invention.

The same reference numerals are used for portions having similar functions and functions throughout the drawings. Throughout the specification, when a part is connected to another part, it includes not only a case where it is directly connected but also a case where the other part is indirectly connected with another part in between. In addition, the inclusion of an element does not exclude other elements, but may include other elements, unless specifically stated otherwise.

Hereinafter, a torque control method using mutual inductance estimation according to an embodiment of the present invention will be described. First, a modeling method of a permanent magnet synchronous motor, a MTPA control not considering mutual inductance, and an MTPA control considering mutual inductance will be described. Then, a mutual inductance estimation method according to an embodiment of the present invention and a torque The control method will be described.

Hereinafter, the modeling of the permanent magnet synchronous motor of the recessed type will be described. The embedded permanent magnet synchronous motor (hereinafter referred to as IPMSM) has a voltage equation as shown in Equation (1).

[Equation 1]

In the equation 1, R is the winding resistance, Ld is the d-axis inductance, Lq is the q-axis inductance, Wr is the rotational angular velocity of the rotor, λm is the flux linkage, Vd, id is the d- Voltage, and current. The equation (1) can be summarized for the current differential term as shown in the following equation (2).

&Quot; (2) "

On the other hand, the magnetic flux can be expressed by the following equation (3).

&Quot; (3) "

Substituting Equation (4) into Equation (2), the following Equation (4) can be obtained, and the current is calculated by integration using Equation (4) and the magnetic flux is obtained using Equation .

&Quot; (4) "

Generally, it is used as a general model in Equations (1) and (3). In this case, since fixed fixed control constants are used, results are different from actual motors. In particular, it is not possible to evaluate the effect of flux saturation or inductance variation. Therefore, a reference table is generally used to consider this. However, it is difficult to analyze the motor mathematically by using the reference table. Therefore, it is difficult to know characteristics of the magnetic flux saturation of the motor because it must be obtained through repeated experiments. As a result, it can be seen that a method for modifying the motor is necessary.

FIG. 1 is a graph showing the FEM analysis results of the d-q axis magnetic flux variation, and FIG. 2 is a graph showing the d-q axis magnetic flux and torque value according to the q axis current. FIG. 1 shows the FEM analysis results of the d and q axis fluxes when only the q-axis current flows. As shown in FIG. 1, it can be seen that as the magnitude of the q-axis current varies, not only the magnetic flux in the q-axis changes but also the d-axis magnetic field. 2, the q-axis magnetic flux saturates while the q-axis current increases, and the d-axis magnetic flux is caused by the dq axis interference component It can be seen that the magnetic flux decreases.

As shown in FIG. 1, it can be seen that the saturation of the magnetic flux and the interference component between the dq axes must be considered. 3 and 4, FEM analysis of the changes of the d-axis magnetic flux component and the q-axis magnetic flux component as the dq-axis current changes, FIG. 3 shows a graph of the d-axis magnetic flux value according to the dq- , And FIG. 4 shows a graph of a q-axis magnetic flux value according to the dq axis current.

As shown in Figs. 3 and 4, interference between each other can be seen in more detail. Accordingly, if the magnetic flux equation is rewritten, it can be expressed as Equation (5) below.

&Quot; (5) "

Where Ldq is the q-axis inductance value for the d-axis, and Lqd is the d-axis inductance value for the q-axis. If this equation is summarized with respect to the mutual inductance and the changing component, the following equation (6) is obtained.

&Quot; (6) "

Equation (6) expresses the term for the fluctuation component separately from Equation (5), and Equation (7) below is an equation that summarizes the fluctuation component and the mutual inductance component into one term. The final expression for the fluctuation component can be expressed by the following equation (8).

&Quot; (7) "

&Quot; (8) "

In Equation _{8 L d0, L q0, λ} m0 is a fixed _{value, L εd (□), L} εq (□) is a variable for the current. This can be expressed by Equation (9) below by converting the equation for the magnetic flux into the equation for the amount of change with respect to the current using the approximation equation for Figs. 3 and 4. [

&Quot; (9) "

In Equation (9), I0 is a d-axis current value having the smallest change amount of the d-axis magnetic flux according to a constant q-axis current change, and ₀ is a d-axis magnetic flux value at this time. K _Ld and K _Lq are values of L _d and L _q when each axis current is 0. K _Sd and K _Sq are constants considering flux saturation degree and K _Sdq and K _Sqd are components for mutual interference.

5 is a graph showing the meaning of the q-axis magnetic flux parameter. That is, the q-axis flux parameter value is obtained as shown in FIG. 5, and the meaning of each parameter can be expressed by the following equations (10) and (11).

&Quot; (10) "

&Quot; (11) "

These equations (10) and (11) can be summarized into the following equations (12) and (13).

&Quot; (12) "

&Quot; (13) "

From Equations (12) and (13), the larger the value of L _Sq , the greater the degree of saturation of the magnetic flux. The larger the value of K _Sqd , the greater the inter-axis interference component.

FIG. 6 is a graph showing the meaning of the d-axis magnetic flux parameter. That is, the d-axis flux parameters are obtained as shown in FIG. 6, and the meaning of each parameter is expressed by the following equations (14) and (15)

&Quot; (14) "

&Quot; (15) "

Using the equations (14) and (15), the values of the equations (16) and (17) are obtained as follows.

&Quot; (16) "

&Quot; (17) "

From Equations (16) and (17), it can be seen that the larger the value of K _Sd , the greater the degree of saturation of the magnetic flux, and the larger the value of K _Sd , the greater the inter-axial interference component and the greater the interval between the graphs. If the magnetic flux equation obtained above is inversely calculated according to the equation for the current, the following equations (18) and (19) for modeling considering magnetic flux saturation can be obtained.

&Quot; (18) "

&Quot; (19) "

Hereinafter, a method of designing the current controller will be described. First, a state feedback type non-interference proportional integral integral controller will be described. Figure 7 shows a block diagram of a typical current PI controller.

The transfer function of the current PI controller can be expressed by the following equation (20).

&Quot; (20) "

If the gain of the PI controller in Equation (20) is set as follows, the transfer function is expressed by Equation (21) below.

&Quot; (21) "

In such a case, the zero point of the PI controller is set to be offset from the pole of the system, and the DC gain of the current controller becomes 1, so that stable and fast response characteristic is shown. After all, the entire current controller is a low-pass filter, the band of the filter is to be determined by the ω _n.

&Quot; (22) "

Figure 8 shows a block diagram of a general PI controller. In actual motor control, there is a limit on the output voltage as described in the inverter for the command voltage. When the integrator of the current controller exceeds the limit voltage, that is, when the integrator saturates due to a sudden increase in the load, it outputs a voltage that is larger than the actual necessary voltage output. This state is called the wind-up of the integrator.

If the integrator saturates, the maximum or minimum value is output as the output voltage irrespective of the input error of the PI controller, and the PI control does not operate properly until the saturation of the integrator disappears. Therefore, an overshoot occurs in the output, and it takes a long time to reach the steady state. In order to prevent such a state of windup, it is necessary to appropriately limit the cumulative error of the integrator. This method is called anti-windup.

There are various methods for this. Recently, most commonly used method is to multiply the difference between the output of the PI controller and the limit value by multiplying the gain of the PI controller by the gain and subtracting it from the integrator . At this time, the gain can be 1 / (3K _P ) ~ 3 / K _P , usually 1 / L _P.

Figure 9 shows a block diagram of a PI controller with an anti-windup function. Figure 10 also shows a block diagram of a general PI controller that does not take into account interference components. As shown in FIG. 10, the conventional PI controller does not compensate for mutual interference components. Therefore, if the motor is moved to the high-speed region, the influence of the motor on the interference component is not taken into consideration, and the overshoot becomes large and the unstable phenomenon in which continuous vibration occurs occurs.

FIG. 11 shows a Bode diagram of the controller while varying the speed of the motor while the control bandwidth of the controller block diagram of FIG. 10 is set to be constant. In this case, the conventional PI controller of FIG. 10, which does not consider the interference components, becomes more unstable as the control moves from the design value toward the high-speed region.

To eliminate the interference component, Fig. 11 shows a block diagram of the PI controller considering the interference component. Equation (23) below is an expression for the mutual interference component of the motor and the counter electromotive force. This compensates for the d-q axis voltage reference. FIG. 12 shows a Bode diagram of a PI controller considering an interference component.

&Quot; (23) "

As a result, it can be seen from the Bode diagram of FIG. 12 that the controller considering the interference component is more stable than the controller which does not consider the mutual interference component.

However, even if a controller that considers mutual interference is used for the above-mentioned typical PI controller, the design value is not satisfied according to the speed. To solve this problem, a cascade PI controller is used. This cascade PI controller can use the inverse model for the controller of the motor to obtain the same value as the current controller design value in all velocity regions.

Figure 14 shows a block diagram of a cascaded PI controller, Figure 15 shows a Bode diagram of a cascaded PI controller, and Figure 16 shows a Bode plot of a cascaded PI controller with a control error of +/- 20% It is.

As shown in FIG. 14, it can be seen that the error of the current is generated in the current regulator part through the PI controller to generate the second current command value. Then, the second current command value is inversely calculated for the interference component in the voltage reference calculator. Then, each gain value can be obtained from the following expression (24).

&Quot; (24) "

As shown in FIG. 15, it can be seen that the design value agrees with the design value in the entire speed region. For the controller, the effect on the parameter error should also be analyzed. Since the value of the inductance of the motor varies with the magnitude of the current and the resistance value changes with temperature, the stability when the parameter error occurs is also an important factor considering the parameter change.

The condition of FIG. 16 is a result of assuming that the resistance, d-axis inductance and q-axis inductance change from -20% to 20%, respectively, under the condition of rotating at a rated speed of 1500RPM. That is, it can be seen that the parameter error is robust.

Hereinafter, abbreviated control will be described. 17 is a graph showing torque, voltage, and current values versus speed in the IPMSM operating region. As shown in FIG. 17, it can be seen that a constant torque region and a constant output region can be classified according to the speed.

The constant torque area means the speed range from the reference speed to the range, usually up to the rated speed. The motor of IPMSM has a reluctance torque because L _ds ≠ L _qs . This relaxation torque can be expressed by the following equation (25).

&Quot; (25) "

To utilize this, both i ^e _ds and i ^e _qs currents are required. Since L _ds <L _qs , when i ^e _ds <0, the reluctance torque is added to the torque by the permanent magnet, so that a larger output torque can be obtained.

)의 크기가 최소가 되는 전류가 존재하는데. 이것은 바로 고정 자 전류 Is를 반지름으로 하는 원과 토크 곡선이 접하는 A점에서의 전류 I_SO이다. 이 전류로 운전하면 단위 전류당 최대 토크가 발생하게 된다. 이와 같이 요구되는 지령 토크 발생을 위해 필요한 고정자 전류가 최소가 되는 i^r _ds와 i^r _qs의 조합을 구해 이에 따라 운전하는 것을 단위 전류당 최대 토 크 제어(Maximum Torque Per Ampere Control)기법이라 한다. Of the myriad current combinations that produce the same magnitude of torque, the stator current Is (=

Lt; RTI ID = 0.0 &gt; of &lt; / RTI &gt; This is the current I _SO at point A where the torque curve is tangent to the circle with the stator current Is as the radius. When operated with this current, the maximum torque per unit current is generated. The combination of i ^r _ds and i ^r _qs that minimizes the stator current required to generate the required command torque and operates accordingly is called the Maximum Torque Per Ampere Control technique.

18 shows a stator current graph for generating the same torque. As shown in FIG. 17, the constant output area corresponds to an area above the reference speed and a weak speed control area.

The IPMSM is driven under a current limiting condition as shown in Equation (26) below and a voltage limiting condition as shown in Equation (27) below.

&Quot; (26) &quot;

&Quot; (27) &quot;

In the high - speed operation region, the abbreviated speed control technique which can generate the maximum torque considering the limit condition of the voltage and current is needed. First, when these two constraints are expressed on the d and q axis current planes, the current limit condition is expressed as a circle with I _SMAX as the radius on the d and q axis current planes. On the other hand, if the voltage limiting condition is represented by a current obtained by substituting the equation (23) into the equation (26), the following equation (28) can be obtained.

&Quot; (28) &quot;

Further, the voltage suggestion ellipse is expressed by the following equation (29).

&Quot; (29) &quot;

FIG. 19 shows a graph of a voltage limiting ellipse and a current suggestion circle. 19 shows the range of the d and q axis command currents i ^{e *} _ds and i ^{e *} _{qs that} can be controlled by a given maximum stator voltage V _smax . As the driving angular velocity _r increases, the voltage required to control the command current of the same magnitude becomes large. However, since the usable voltage V _smax is fixed, the area of the command current that can be controlled according to the angular speed? _R becomes small. The decrease in the elliptical area as the speed increases means this.

When operating the IPMSM, both of these voltage and current limitations must be satisfied. The area of controllable command current when both of these constraints are considered is a common area between the voltage limit ellipse and the current limiting source. Therefore, the command current should always be given to the inside of the region to be controlled.

Figure 20 shows an overall block diagram using a voltage regulator. The method shown in Fig. 20 is a kind of abbreviation control technique generally used by the voltage regulation method. This abbreviation control method is a control method that is used in the form of adding the value after the controller to the d-axis current reference with the maximum voltage of the stator and the voltage of the output to the output.

The advantage of this technique is that it is insensitive to parameter variations because it does not use parameter values of the motor. However, the disadvantage is that the durability is low and the stator voltage is not used to the maximum. Among the disadvantages, the problem of the dependency is that the difference of the voltage does not directly affect the voltage output value but the influence of the dc axis current reference, do.

Hereinafter, a torque control method using mutual inductance estimation according to an embodiment of the present invention will be described. First, the MTPA control that does not consider the mutual inductance and the MTPA control that considers the mutual inductance are compared and the mutual inductance estimation method according to one embodiment of the present invention will be described.

)로 정규화하면 이하의 수학식 30으로 표현될 수 있다. When an arbitrary command torque is given, the minimum stator current for MTPA control is obtained and used. Criteria as the above-mentioned torque and current in equation (25) follows the current ib (= φf / (Lqs- Lds) and the reference torque T _b (=

) Can be expressed by the following equation (30).

&Quot; (30) &quot;

From this, the normalized torque is expressed by the rectified currents as shown in Equation 31 below.

&Quot; (31) &quot;

The process of finding the minimum current for MTPA control is as follows. First, when the command torque of T _en is given, try to find the minimum d-axis current i _dno . First, the stator current _limiting formula is expressed by a function of i _dno using equation (31).

(32)

This equation (32) is differentiated by i _dno and is set to 0 as follows, and the solution of the equation is obtained, so that the minimum d-axis current for torque generation of a given T _en can be obtained.

&Quot; (33) &quot;

A similar procedure is used to obtain the minimum q-axis current iqno for the generation of the command torque Ten given. The current limit equation expressed by iqno, which is expressed by iqno, is differentiated by iqno and set to 0, which yields a quadratic equation as shown in Equation (35) below. When this is solved, the minimum q- .

&Quot; (34) &quot;

&Quot; (35) &quot;

In order to control the maximum torque per unit current with respect to the required output torque command, the d-q axis current command must be obtained from the equations (33) and (35). Fig. 21 shows an optimum current command graph according to the command torque.

If the required output torque command is given by the stator current command I _s , the current command of the dq axis for maximum torque control per unit current should be obtained. First, the torque of Equation (25) is expressed by stator current I _s , which corresponds to Equation (36) below.

&Quot; (36) &quot;

Where β is the angle between the d ^r axis and the stator current I _s . If it is differentiated as shown in Equation 37 and then solved by 0,

&Quot; (37) &quot;

&Quot; (38) &quot;

It is possible to obtain the current angle beta that generates the maximum torque per unit current as shown in Expression (38). From this, the optimal current i ^r _dso = I _{s cos β} , i ^r _qso = I _s sin β is obtained.

axis current and the d-axis current with respect to the q-axis current, the MTPA equation can be expressed by the following equation (39).

[Equation 39]

However, the MTPA method which does not take into account the mutual inductance mentioned above is changed in value by the change of the inductance. So, instead of calculating the above method each time, it is obtained offline in advance and used as a look-up table. However, there is a drawback that the reference table method can not be used without an offline experiment. Therefore, if MTPA control is performed again considering the mutual inductance, the calculated value is more accurate than the formula that does not take into consideration the mutual inductance.

If the voltage equation is rewritten in consideration of the mutual inductance, it can be expressed by the following equation (40).

[Equation 40]

Then, if the torque equation is rewritten using the equation (40), it becomes equal to the equation (41).

(41)

Here, expressing the portion represented by the d-q axis current as the magnitude and?

(42)

In order to find the maximum torque point in Equation (42), if?

Equation (43)

Also, as mentioned above, i _d = Iscos 硫 and i _q = I _s sin 硫. The equation derived from the equation (43) can be rearranged and expressed by the following equation (44).

&Quot; (44) &quot;

The process of finding the maximum point using Equation (44) is expressed by Equation (45) below.

&Quot; (45) &quot;

&Quot; (46) &quot;

Equation (46) is an MTPA type in which mutual inductance is taken into consideration, and it can be seen that there is a difference from Equation (39) which does not take into account the above mutual inductance. In addition, parameter values used in the equation 46 is any number other than the L _{εd, εq} L is a fixed parameter, and corresponds to the value of the variable L _εd, L _εq.

Hereinafter, a mutual inductance estimation method according to an embodiment of the present invention will be described. To estimate the mutual inductance, the inductance nonlinear characteristics of the motor should be considered. The nonlinear characteristics can be summarized by Equation (47) as follows: Equation (1) and Equation (40) are used to rearrange all the terms of the mutual interference component in terms of the back electromotive force.

&Quot; (47) &quot;

Here, the counter electromotive force component is expressed by the following equation (48).

&Quot; (48) &quot;

As shown in Expression 47, when all the interference components are included in the back electromotive force component, the following Expression 48 is obtained. This back electromotive force component is referred to as an extended induced voltage, and when the extended induced voltage is estimated, a conclusion can be drawn about the mutual interference component.

That is, if the extended induced voltage of Equation (48) is estimated, it can be rewritten as an interference component.

&Quot; (49) &quot;

or

(50)

or

The inductance values of the mutual interference components obtained by using the estimated extended electromotive voltage are expressed by Equations (49) and (50). As a result, these extended induced voltages must be estimated.

The formula for estimating the extended organic voltage is summarized as follows.

&Quot; (51) &quot;

Equation (51) is an equation summarized for the d-axis current in the general voltage equation. Using this equation (51), a virtual current model is created.

(52)

Equation (52) is the same as the equation shown in Equation (51), the voltage and the q-axis current use the same value, and the extended organic voltage component uses the virtual model. As a result, the value of the extended induced voltage is estimated using Equations (51) and (52).

&Quot; (53) &quot;

As a result, when the error converges to 0 through the PI controller, the value of the extended induced voltage can be obtained.

(54)

Equation (54) proceeds in the same manner as Equation (51), which corresponds to a formula for estimating the extended induced voltage value of the q-axis using the q-axis current and the voltage.

(55)

The values of the model current and the model extension induced voltage are as shown in Equation (55). As a result, the value of the extended induced voltage corresponds to Equation (56) below.

&Quot; (56) &quot;

FIG. 8 shows a control block on the basis of equations (53) and (56). The mutual inductance value is estimated using Equation (49) and Equation (50) using the extended induced voltage estimated in Fig.

Hereinafter, the torque control method using the aforementioned mutual inductance estimation will be described. Among the proposed methods for controlling the permanent magnet motor, there is a direct torque control method. The goal of the direct torque control method is to combine the simplicity of the controller configuration, which is the advantage of the V / F control method, and the robustness of the motor control constant to the excellent torque control performance, which is an advantage of the vector control method.

This method was first proposed by Takahashi and Noguchi of Japan in 1984 by Depenbrock in Germany in 1985, almost the same time, and was applied to early induction motors and then applied to permanent magnet type motors by Rahman and Morimoto .

A typical control method of the direct torque control method will be described. In controlling the stator flux and torque, a proper inverter voltage vector is selected and output in order to minimize the control error by comparing the actual value and the set value at every sampling period. The magnetic flux and the torque are directly controlled without using the PWM method. In principle, in order to increase the torque, in order to increase the frequency of the effective voltage vector to output and decrease, the frequency of the zero voltage vector is increased to reduce the torque. The magnitude of the magnetic flux is controlled by selecting a voltage vector that increases or decreases the magnetic flux among the effective voltage vectors adjacent to the stator magnetic flux vector. The magnitude of the output voltage and current is indirectly controlled.

The direct torque control method has an advantage that the response of the torque is very fast because the inverter voltage vector is determined only by the spatial position of the hysteresis comparator and the stator flux at every cycle and the voltage is output without delay by the controller. Therefore, the response characteristics against abrupt load fluctuations and instantaneous command fluctuations are excellent. At this time, the hysteresis comparator compares the magnitude between the command value and the actual value, and the magnitude of the flux and torque ripple is determined by the hysteresis amplitude.

The direct torque control method has a simpler structure than the vector control method and does not require the coordinate axis conversion from the stator coordinate system to the rotor coordinate system. The direct torque control method has a small dependence on the motor control constant, Since the magnetic flux can be controlled only by the absolute position of the magnetic flux and the information of the position belonging to which sector among the six sectors defined at the intervals of 60 degrees, accurate magnetic flux information is not required compared with the vector control.

Therefore, the DTC can be implemented with comparatively simple hardware, and the torque control characteristic is advantageous in that it is not greatly influenced by the change of the motor control constant. In order to utilize the advantage of the direct torque control method as described above, there are various problems to be improved.

That is, it is easy to generate a large torque ripple at the time of starting or at a low speed, and the torque ripple is relatively large in the other regions as compared with the PWM system. In addition, there is a problem that switching frequency, which is one of the important variables in the design of the inverter, is greatly influenced by the amplitude of the hysteresis comparator and varies with other factors. In particular, although the switching frequency is controlled according to the amplitude of the hysteresis comparator, the switching frequency varies depending on the operation frequency and load condition of the motor even if the same amplitude is maintained and operated.

In the normal torque control method, like the DTC, the switching function is tabulated and the current command value is compensated by using the reference table method. However, there are many restrictions on the load test of all the motors, and it is necessary to compensate the error by many experiments.

Accordingly, in one embodiment of the present invention, the torque value is calculated using the estimated mutual inductance value and torque control is performed.

In order to control the torque at the estimated mutual inductance value according to an embodiment of the present invention, the torque inequality should be rearranged into a torque equation considering mutual inductance components. In order to do so, the following equation (57) is summarized, and the mechanical energy portion in the total energy becomes the third term, and the torque energy is expressed by the following equation (58).

&Quot; (57) &quot;

Equation (58)

The torque obtained by the equation (58) is used to control the torque. When I controller is constructed using the estimated torque, there is a problem in responsiveness. Therefore, in order to solve this problem, the forward compensation is performed through the equation (25) where the mutual inductance is not considered. This means that the I controller is used to compensate for the torque error only for the torque portion due to mutual inductance. 9 shows a torque control block diagram showing the configuration of the I controller.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. It is to be understood that such modified embodiments are within the scope of protection of the present invention as defined by the appended claims.

Claims (4)

An apparatus for controlling torque of a permanent magnet synchronous motor (IPMSM)
A mutual inductance estimating unit for estimating a d-axis mutual inductance value and a q-axis mutual inductance value according to the torque command value;
A torque value estimator for estimating a torque control value based on the mutual inductance estimated value estimated by the mutual inductance estimating unit;
A dq axis current command value generator for generating a d axis current command value and a q axis current command value based on the estimated torque control value estimated by the torque value estimator; And
And a current controller for controlling the torque of the IPMSM by applying a current value to the IPMSM based on the current command value generated by the dq axis current command value generator.
The method according to claim 1,
Wherein the current command generator generates a current command value by the following Equation (1) by applying a maximum torque (MTPA) control technique per unit current in consideration of mutual inductance: &lt; EMI ID = 1.0 &gt;
[Equation 1]

3. The method of claim 2,
Wherein the mutual inductance estimation value is determined by the following equation (2) or (3): &lt; EMI ID = 1.0 &gt;
&Quot; (2) &quot;

&Quot; (3) &quot;

The method of claim 3,
Wherein the torque estimation value is determined by the following equation (4): &lt; EMI ID = 4.0 &gt;
[Equation 4]

Images