JP4760118B2 - Electric motor control device - Google Patents

Description

  The present invention relates to a motor control device, and more particularly to a technique for compensating for dead time in an inverter circuit.

As a conventional technique for compensating for dead time, there is a technique described in Patent Document 1 (Japanese Patent Laid-Open No. 61-231889). This conventional example describes a configuration in which the dead time compensation amount is changed based on the three-phase current value.
Further, in the conventional example described in the following Patent Document 2 (Japanese Patent Laid-Open No. 6-98595), a configuration is described in which the dead time compensation amount is changed based on an average value after the absolute value of the three-phase current value. .

JP 61-231889 A JP-A-6-98595

In Patent Document 1, as a method for obtaining an optimum dead time compensation amount for a three-phase current value, a direct current is passed through a stopped motor, and a dead time compensation amount is calculated from a difference between a voltage command value and an actual voltage at that time. Is described. However, when a small DC current is passed through the stopped motor, the voltage becomes a very small value, and there is a problem that the compensation amount cannot be obtained correctly due to a measurement error. Further, in the method of combining the voltage command value and the actual voltage, it is necessary to obtain the compensation amount for each carrier frequency when the carrier frequencies are different. In addition, even if it is obtained to some extent correctly, problems such as compensation delay are added when the motor is rotating, and the d-axis voltage command value and the q-axis voltage command value change when the carrier frequency is switched. There is. It is difficult to obtain a compensation amount that does not change the d-axis voltage command value and the q-axis voltage command value because the three-phase current and the three-phase voltage are AC. When the d-axis voltage command value and the q-axis voltage command value change, there arises a problem that the current changes and torque fluctuation occurs.
Further, in Patent Document 2, since the current value during rotation is represented by one value, when the carrier frequency is switched while the electric motor is rotating, the d-axis voltage command value and the q-axis voltage command value are changed. It is only necessary to obtain the compensation amount for preventing the change as one value, and the adjustment can be easily performed. However, the average value after the absolute value of the three-phase current value requires long time data in order to stabilize the calculated value in the same manner as the effective value of the three-phase current value, and therefore the calculation amount increases. When the three-phase current value suddenly changes, there is a problem that a delay occurs and the current control is adversely affected.
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide an electric motor control device that is easy to adjust and can perform highly accurate dead time compensation in the entire operation region. .

To achieve the above object, in claim 1 of the present invention, the dead time compensation means is determined by the equation _{i a} = √ from d-axis current _{i d} and the q-axis current ^{_{i q (i d 2 + i}} q 2) is configured to set the compensation amount of dead time based on the size i _a of the current vector.

  According to the present invention, there is an effect that there is no problem of delay and the control becomes stable as compared with the conventional case where the compensation amount is calculated based on the effective value of the three-phase phase current. In addition, when the carrier frequency is switched while the electric motor is rotating, the current is not changed, and as a result, the torque fluctuation can be prevented from occurring.

Example 1
FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention. This embodiment shows a configuration in which dead time compensation is performed based on an estimated current value.
In FIG. 1, the current map unit 1 calculates the d-axis from the torque command value T ^* [N · m] given from the outside and the rotational speed [rotor angular velocity ω (electrical angle) [rad / s] of the motor 7]. The current target value i _d ^* [A] and the q-axis current target value i _q ^* [A] are obtained by map drawing.

The torque command value T ^* is a signal that indicates the output torque of the electric motor 7, and is set according to, for example, the amount of depression of the accelerator pedal. Further, the rotor angular velocity ω of the electric motor 7 is obtained by differentiating the rotor phase (electrical angle) θ [rad] of the electric motor 7 obtained by a position detector such as a resolver or an encoder by the differentiating unit 9.

The current control unit 2 includes the d-axis current target value i _d ^* and the q-axis current target value i _q ^*, and a d-axis current value i _d [A] and a q-axis current value i _q [A] described later. In accordance with the deviation, the d-axis voltage command value v _d ^* [V] and the q-axis voltage command value v _q ^* [V] are calculated by the following equation (1).

ただし、
ｖ_ｄ ^＊：ｄ軸電圧指令値［Ｖ］ ｉ_ｄ：ｄ軸電流値［Ａ］
ｖ_ｑ ^＊：ｑ軸電圧指令値［Ｖ］ ｉ_ｑ：ｑ軸電流値［Ａ］
Ｋ_ｐｄ：ｄ軸比例ゲイン Ｋ_ｐｑ：ｑ軸比例ゲイン
Ｋ_ｉｄ：ｄ軸積分ゲイン Ｋ_ｉｑ：ｑ軸積分ゲイン
ｓ：ラプラス演算子
また、この際、必要に応じて非干渉制御を適用しても良い。
なお、以下の説明において、ｄ軸とｑ軸をまとめて表現する場合には、ｄ−ｑ軸電圧指令値やｄ−ｑ軸電流値のように略して表示する場合もある。

However,
v _d ^* : d-axis voltage command value [V] i _d : d-axis current value [A]
v _q ^* : q-axis voltage command value [V] i _q : q-axis current value [A]
K _pd : d-axis proportional gain K _pq : q-axis proportional gain K _id : d-axis integral gain K _iq : q-axis integral gain s: Laplace operator In this case, even if non-interference control is applied, good.
In the following description, when the d-axis and the q-axis are expressed together, they may be abbreviated and displayed as a dq-axis voltage command value or a dq-axis current value.

The three-phase converter 3 performs a two-phase to three-phase conversion based on the following (Equation 2) to perform the d-axis voltage command value v _d ^* [V] and the q-axis voltage command value v _q ^* [V]. Is converted into a three-phase voltage command value, that is, a u-phase voltage command value v _u ^* [V], a v-phase voltage command value v _v ^* [V], and a w-phase voltage command value v _w ^* [V]. At this time, the rotor phase θ of the electric motor 7 may be a value compensated for delay.

ただし
ｖ_ｕ ^＊：ｕ相電圧指令値［Ｖ］ ｖ_ｖ ^＊：ｖ相電圧指令値［Ｖ］
ｖ_ｗ ^＊：ｗ相電圧指令値［Ｖ］ ｖ_ｄ ^＊：ｄ軸電圧指令値［Ｖ］
ｖ_ｑ ^＊：ｄ軸電圧指令値［Ｖ］ θ：電動機７の回転子位相（電気角）［ｒａｄ］。

However, _vu ^* : u-phase voltage command value [V] _vv ^* : v-phase voltage command value [V]
v _w ^* : w-phase voltage command value [V] v _d ^* : d-axis voltage command value [V]
v _q ^* : d-axis voltage command value [V] θ: rotor phase (electrical angle) [rad] of the motor 7.

The PWM conversion unit 4 performs PWM conversion based on the following (Equation 3), and calculates the u-phase pulse width t _u [s], the v-phase pulse width t _v [s], and W from the three-phase voltage command value. The phase pulse width t _w [s] is calculated.

ただし、
Ｔ_０：ＰＷＭキャリア周期［ｓ］ ｖ_ｄｃ：直流電圧［Ｖ］
ｔ_ｕ：ｕ相パルス幅［ｓ］（補償前） ｔ_ｖ：ｖ相パルス幅［ｓ］（補償前）
ｔ_ｗ：Ｗ相パルス幅［ｓ］（補償前）
なお、ＰＷＭキャリア周期Ｔ_０はＰＷＭ変換する際に用いるキャリア信号の周期であり、直流電圧ｖ_ｄｃは電動機７を駆動する直流電源の電圧である。

However,
T ₀ : PWM carrier cycle [s] v _dc : DC voltage [V]
t _u : u-phase pulse width [s] (before compensation) t _v : v-phase pulse width [s] (before compensation)
t _w : W-phase pulse width [s] (before compensation)
Note that the PWM carrier cycle T ₀ is the cycle of the carrier signal used for PWM conversion, and the DC voltage v _dc is the voltage of the DC power source that drives the motor 7.

The dead time compensation unit 5 obtains a compensation amount determined according to the three-phase current values (i _u , i _v , i _w ) flowing through the electric motor 7 and adds them to the respective pulse widths of each uvw phase. The u-phase pulse width t _u ′ [s], the v-phase pulse width t _v ′ [s], and the W-phase pulse width t _w ′ [s] are calculated.
However, the three-phase current values used in this calculation may be actual current values i _u , i _v , i _w , or three-phase currents obtained by converting the dq-axis current target values i _d ^* , i _q ^* into three phases. The target values i _u ^* , i _v ^* , and i _w ^* may be used, and the dq axis current estimated value obtained by applying a filter corresponding to the current response to the dq axis current target values i _d ^* and i _q ^*. Three-phase current estimated values i _u ^, i _v ^, i _w ^ obtained by three-phase conversion may be used. Details of this will be described later.

The inverter 6 operates in response to the PWM signal having the u-phase pulse width t _u ′, the v-phase pulse width t _v ′, and the W-phase pulse width t _w ′ after the dead time compensation, and the power of a DC power supply (not shown) _Is converted into a three-phase AC voltage (v _u , v _v , v _w ) to drive the electric motor 7.
However, v _{u is the} u-phase voltage value [V], v _{v is the} v-phase voltage value [V], and v _{w is the} w-phase voltage value [V]. These are hereinafter referred to as actual voltages.
At this time, currents i _u [A], i _v [A], and i _w [A] flowing in the u, v, and w phases of the electric motor 7 are detected by a current sensor (not shown). Further, the rotor phase (electrical angle) θ [rad] of the electric motor 7 is detected by using a position detector such as a resolver or an encoder (not shown).

The dq axis conversion unit 8 converts the three phase current values from the detected three phase currents i _u , i _v , i _w and the rotor phase θ into the dq axis current based on the following equation (4). Convert to values i _d , i _q .

上記のｄ−ｑ軸電流値ｉ_ｄ、ｉ_ｑが前記（数１）式のｄ軸電流値ｉ_ｄ、ｑ軸電流値ｉ_ｑとして用いられる。
なお、上記の電流マップ部１、電流制御部２、三相変換部３、ＰＷＭ変換部４、デッドタイム補償部５、ｄ−ｑ軸変換部８、微分部９および後記電流制御相当フィルタ１０と三相変換部１１、１１’の部分は、ＣＰＵとメモリおよびそれらに付随する電子回路で構成することが出来る。

The d-q axis current values i _d and i _q are used as the d-axis current value i _d and the q-axis current value i _{q in the} equation (1).
The current map unit 1, current control unit 2, three-phase conversion unit 3, PWM conversion unit 4, dead time compensation unit 5, dq axis conversion unit 8, differentiation unit 9, and current control equivalent filter 10 described later The three-phase converters 11 and 11 ′ can be constituted by a CPU, a memory, and an electronic circuit associated therewith.

Hereinafter, the dead time compensation which is the gist of the present invention will be described.
In FIG. 1, the current control equivalent filter 10 and the three-phase conversion are used in order to obtain the three-phase current estimated values i _u ^ [A], i _v ^ [A], i _w ^ [A] used in the dead time compensation unit 5. Part 11 is provided.
First, the current control equivalent filter 10 applies a filter equivalent to current control in the current control unit 2 to the d-axis current target value i _d ^* [A] and the q-axis current target value i _q ^* [A] to estimate the d-axis current. A value i _d ^ [A] and a q-axis current estimated value i _q ^ [A] are calculated. The three-phase conversion unit 11 performs two-phase to three-phase conversion on the d-axis to q-axis current estimation values (i _d ^, i _q ^) to obtain a three-phase current estimation value, that is, a u-phase current. Estimated value i _u ^ [A], v-phase current estimated value i _v ^ [A], and w-phase current estimated value i _w ^ [A] are calculated. Further, the d-axis current estimated value i _d ^ and the q-axis current estimated value i _q ^ obtained by the current control equivalent filter 10 are used when the d-axis current value i is calculated when _a current vector magnitude ia to be described later is obtained. _d, it can also be used in place of the _{i q.}
In the first embodiment, the dead time compensation unit 5 performs dead time compensation using the three-phase current estimation values i _u ^, i _v ^, and i _w ^ (details will be described later).

(Example 2)
FIG. 2 is a block diagram showing a configuration of the second embodiment of the present invention, and is a block diagram in a case where dead time compensation is performed based on a target current value.
In FIG. 2, the three-phase conversion unit 11 ′ performs two-phase to three-phase conversion on the d-axis current target value i _d ^* and the q-axis current target value i _q ^* to obtain a three-phase current target value, that is, a u-phase current target value. i _u ^* [A], v-phase current target value i _v ^* [A], w-phase current target value i _w ^* [A] are calculated, and these three-phase current target values i _u ^* , i _v ^* , i _The dead time compensation unit 5 performs dead time compensation using _w ^* (details will be described later).

FIG. 3 is a block diagram showing a configuration of the third embodiment of the present invention, and is a block diagram in a case where dead time compensation is performed based on an actual current value.
In FIG. 3, dead time compensation is performed in the dead time compensation unit 5 using the detected three-phase current values i _u , i _v , i _w of the electric motor 7 (details will be described later).

Hereinafter, the dead time compensation in the dead time compensation unit 5 will be described in detail.
(Example 1 of dead time compensation)
FIGS. 4 and 5 are block diagrams showing the first embodiment of the dead time compensation unit 5 in FIGS.
Figure 4 is a configuration in which do not change the carrier frequency f _c in the PWM converter section 4, FIG. 5 is a configuration for changing the carrier frequency f _c.

4 and 5, t _u , t _v , and t _w are pulse widths of u, v, and w phases before dead time compensation, and t _u ′, t _v ′, and t _w ′ are after dead time compensation. The pulse width of each phase of u, v, and w. Also, 12 and 13 is a compensation amount map storing the relationship between the compensation amount a of current vector i _a and the dead time, one in FIG. 4 without changing the carrier frequency f _c only, and changes the carrier frequency f _c in Figure 5, it is provided plural in accordance with the carrier frequency f _c. Reference numerals 14, 15 and 16 denote polarity setting units for each phase, and select either positive or negative depending on the magnitude relationship between the currents i _u , i _v , i _{w of} each phase and the compensation amount a. To do. Also, the magnitude i _a of the current vector is obtained from the dq-axis current values i _d and i _q by the formula i _a = √ ( _id ² + i _q ² ) [A].
As described above, FIG. 4, 5, the compensation amount a is set according to the size i _a of the current vector, the three-phase currents i _u, i _v, i _w is only by switching the polarity of the compensation Use.

In FIGS. 4 and 5, the three-phase currents i _u , i _v , and i _w are shown, but these may be actual currents i _u , i _v , and i _w , or the three-phase currents may be used. Estimated values i _u ^, i _v ^, i _w ^ may be used, or three-phase current target values i _u ^* , i _v ^* , i _w ^* may be used. For example, in the case of FIG. 1, the three-phase current estimated values i _u ^, i _v ^, i _w ^ are used, and in the case of FIG. 2, the three-phase current target values i _u ^* , i _v ^* , i _w ^* are used. In the case of FIG. 3, real currents i _u , i _v and i _w are used. Moreover, d-q axis current values _i d, _{i q} used for the operation of the current vector _{i a,} the actual d-q axis current values _i d, may be the _{i q,} d-q axis current target value _i ^{d *} , I _q ^* , or dq axis current estimation values i _d ^, i _q ^.

The compensation amount maps 12 and 13 are created by a method in which the voltage command values (v _u ^* , v _v ^* , v _w ^* ) and the actual voltages (v _u , v _v , v _w ) are set to coincide with each other. Alternatively, a method in which the relationship between the voltage command value and the actual voltage does not change even when the carrier frequency is switched (not equal to the actual voltage) may be used. These maps may be created using values obtained through experiments in advance, or approximate mathematical expressions may be used instead of maps.

  The method of setting the compensation amount to match the voltage command value and the actual voltage has an effect that stable operation can be performed because no error is added to the constant of the motor 7 to be controlled. In addition, the method of preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched has the effect of simplifying the configuration and reducing the memory capacity and calculation load in a system that switches the carrier frequency. is there.

Further, in the case of setting the compensation amount so that the voltage command value and the actual voltage coincide with each other, the configuration shown in FIG. 4 may be used unless the carrier frequency is changed, and the configuration shown in FIG. 5 is used when the carrier frequency is changed.
In the method of preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched, the configuration of FIG. 4 may be used even when the carrier frequency is changed.

The current vector magnitude i _a , u-phase current value i _u , v-phase current value i _v , and w-phase current value i _w are the current target values i _u ^* even if the actual current values i _u , i _v , and i _w are used ^. , I _v ^* , i _w ^* , and current estimated values i _u ^, i _v ^, i _w ^ estimated from the target current value (obtained by applying a filter corresponding to the current response to the target current value) However, it is more stable if the actual current value is not used. Further, when the current control operates stably, the current estimated value is operated more stably than the current target value is used.

The reference compensation amount maps 12, 13 may be used torque command values T ^* instead of size i _a current vector, but the torque estimated value T ^ was filtered torque command value T ^* good.
When there is a dependency on the rotational speed, a map for each rotor angular speed ω (electrical angle) of the electric motor 7 may be used.
As an example, as shown in FIG. 12, when the current increases from 0 to a constant value, the amount of compensation increases linearly, and then becomes a constant value, that is, approximated by a combination of simple straight lines, the map is changed. A simple linear expression is stored in the memory, and the compensation amount can be obtained using the stored linear expression, so that the memory capacity and the calculation amount can be reduced.

  With the configuration described above, in the first embodiment of dead time compensation, adjustment is easy even when the motor 7 is rotating, and the current effective value is used (conventional example: Japanese Patent Laid-Open No. Hei 6 (1994)). Compared to Japanese Patent No. 98595), there is an effect that there is no delay caused by calculation.

(Example 2 of dead time compensation)
FIGS. 6 and 7 are block diagrams showing a second embodiment of the dead time compensation unit 5 in FIGS. The circuits shown in FIGS. 6 and 7 are configured so that a basic compensation amount is set according to the three-phase current, and a value obtained by correcting the compensation amount according to the magnitude of the current vector is used as the compensation amount. Is.
Figure 6 is a configuration in which do not change the carrier frequency f _c in the PWM converter section 4, FIG. 7 is a configuration for changing the carrier frequency f _c.

6 and 7, t _u , t _v and t _w are the pulse widths of u, v and w phases before dead time compensation, and t _u ′, t _v ′ and t _w ′ are after dead time compensation. The pulse width of each phase of u, v, and w. Reference numerals 17 to 19 and 17 ′ to 19 ′ are compensation amount maps storing the relationship between the current values i _u , i _v , and i _w of each phase and the compensation amounts, and 20 and 20 ′ are the magnitudes of the current vectors. i is a correction amount map that stores the correction amount by _a. Only one in every 6 each phase current does not change the carrier frequency f _c, in Figure 7 to change the carrier frequency f _c, are provided plural in accordance with the carrier frequency f _c. 21 to 23 are variable gain multipliers, and K is an adjustment gain. In the variable gain multipliers 21 to 23, the value of the adjustment gain K changes according to the correction value read in the maps 20, 20 ′, and a value obtained by multiplying the value sent from the compensation amount map by the adjustment gain K is output. It is supposed to be.

As described above, in the configurations of FIGS. 6 and 7, the basic compensation amounts are obtained from the current values i _u , i _v , i _{w of the} respective phases using the compensation amount maps 17 to 19 and 17 ′ to 19 ′. , it is configured to be corrected based on the compensation amount to the magnitude i _a of the current vector.
The above current values _{i u,} i v, as the _{i w,} as described above, the actual current _{i u,} i v, may be the _{i w,} three-phase current estimated value _{i u ^, i v ^,} i _It may be _w ^ or a three-phase current target value i _u ^* , i _v ^* , i _w ^* . For example, in the case of FIG. 1, the three-phase current estimated values i _u ^, i _v ^, i _w ^ are used, and in the case of FIG. 2, the three-phase current target values i _u ^* , i _v ^* , i _w ^* are used. In the case of FIG. 3, real currents i _u , i _v and i _w are used.

Also, if there is a rotational speed dependent on the correction value based on the magnitude i _a of the current vector may be a map 20, 20 'as a value for each rotational speed change.
Each map may be created by a method of setting a compensation amount that matches the voltage command value and the actual voltage, or a method for preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched. (It is not equal to the actual voltage). These maps may be created using values obtained by experiments in advance, or approximate mathematical formulas may be used.

  The method of setting the compensation amount to match the voltage command value and the actual voltage has an effect that stable operation can be performed because no error is added to the constant of the motor 7 to be controlled. In addition, the method of preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched has the effect of simplifying the configuration and reducing the memory capacity and calculation load in a system that switches the carrier frequency. is there.

Further, in the case of setting the compensation amount so that the voltage command value and the actual voltage coincide with each other, the configuration shown in FIG. 6 may be used unless the carrier frequency is changed, and the configuration shown in FIG. 7 is used when the carrier frequency is changed.
In the method in which the relationship between the voltage command value and the actual voltage does not change even when the carrier frequency is switched, the configuration of FIG. 6 may be used even when the carrier frequency is changed.

In addition, as described above, the current vector magnitudes i _a , u-phase current values, v-phase current values, and w-phase current values may be actual current values, current target values, or current estimation values estimated from current target values. However, a combination thereof may be used, but the operation without using the actual current value is more stable.

Further, when the current control operates stably, the current estimated value is operated more stably than the current target value is used.
To see the map may be using the torque command value T ^* instead of i _a, may be using the torque estimated value T ^ multiplied by the filter to the torque command value T ^*.

  With the configuration as described above, adjustment during rotation is difficult in the configuration using the three-phase current, but in the second embodiment of dead time compensation, there is an effect that adjustment is easy even during rotation.

(Embodiment 3 of dead time compensation)
FIGS. 8 and 9 are block diagrams showing a third embodiment of the dead time compensation unit 5 in FIGS. The circuits shown in FIGS. 8 and 9 use a configuration that uses a three-phase current when the motor is stopped and rotates at a low speed, and switches so that the configuration described in the first embodiment of dead time compensation is used when the motor rotates at a medium and high speed. is there.
Figure 8 is a configuration in which do not change the carrier frequency f _c in the PWM converter section 4, FIG. 9 is a configuration for changing the carrier frequency f _c.

8 and 9, t _u , t _v , and t _w are the pulse widths of u, v, and w phases before dead time compensation, and t _u ′, t _v ′, and t _w ′ are after dead time compensation. The pulse width of each phase of u, v, and w. 12, 13 and 14 to 16 are the same as those in FIGS. 4 and 5, and 17 to 19, 17 ′ to 19 ′ and 20 and 20 ′ are the same as those in FIGS. 6 and 7. is there.

Reference numerals 24 to 26 denote switching units, which are switched to the maps 17 to 19 and 17 ′ to 19 ′ in the stop state or at the low speed rotation according to the rotation speed (rotor angular speed ω) of the electric motor 7, and are medium to high speed. When rotating, it is switched to the 14-16 side.
In addition, about the rotational speed which switches the switchers 24-26, the structure (17-19 side) using a three-phase electric current has a difference | error which is difficult to adjust during middle-high speed rotation, and uses the magnitude | size of a current vector (14- 16 side) has a slight error in principle at stop and low speed, and the most advantageous rotation speed utilizing each method is determined by experiment.

  Each map may be created by a method of setting a compensation amount that matches the voltage command value and the actual voltage, or a method for preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched. (It is not equal to the actual voltage). These maps may be created using values obtained through experiments in advance, or approximate mathematical expressions may be used instead of maps.

  The method of setting the compensation amount to match the voltage command value and the actual voltage has an effect that stable operation can be performed because no error is added to the constant of the motor 7 to be controlled. In addition, the method of preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched has the effect of simplifying the configuration and reducing the memory capacity and calculation load in a system that switches the carrier frequency. is there.

Further, in the case of the method of setting the compensation amount to match the voltage command value and the actual voltage, the configuration of FIG. 8 may be used unless the carrier frequency is changed, and the configuration of FIG. 9 is used when the carrier frequency is changed.
In the method of preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched, the configuration of FIG. 8 may be used even when the carrier frequency is changed.

  As described above, the magnitude of the current vector, the u-phase current value, the v-phase current value, and the w-phase current value may be the actual current value, the current target value, or the current estimated value estimated from the current target value. However, it is more stable if the actual current value is not used.

When the current control operates stably, the current estimated value is operated more stably than the current target value is used.
In addition, reference maps may be the torque command value T ^* instead of i _a, may be torque estimated value T ^ multiplied by the filter to the torque command value T ^*.
When there is a dependency on the rotational speed, a map for each rotor angular speed ω (electrical angle) of the electric motor 7 may be used.

  Further, in the configuration using the three-phase current (on the side of 17 to 19), for example, as shown in FIG. 13, for example, the compensation amount increases linearly from 0 to a constant interval, and before and after that, a constant value In the configuration using the magnitude of the current vector or using the magnitude of the current vector (on the side of 14 to 16), as shown in FIG. , A simple linear expression can be stored in the memory instead of the map, and the compensation amount can be obtained using the stored linear expression, so that the memory capacity and the calculation amount can be reduced. In other words, the memory capacity and the calculation amount can be reduced by approximating with simple straight line combinations.

  As described above, in the third embodiment of the dead time compensation, the configuration using the three-phase current is difficult to adjust during the middle / high speed rotation and has an error, and the configuration using the magnitude of the current vector (the first embodiment) is stopped / Since there is some error in principle at low speed, the combination of these two, the configuration using the three-phase current is used in the stopped state / low-speed rotation, and the configuration using the magnitude of the current vector is otherwise ( By using it for medium and high speed rotation), it is possible to perform highly accurate dead time compensation in the entire operation region.

(Embodiment 4 of dead time compensation)
FIGS. 10 and 11 are block diagrams showing a fourth embodiment of the dead time compensation unit 5 in FIGS. 10, the circuit shown in Figure 11, sets the basic compensation amount based on the three-phase current, construction of the compensation amount correction value a compensation amount of the in accordance with the magnitude i _a of the current vector (Embodiment 2 of dead time compensation) is combined with a configuration in which the correction value is switched in accordance with the rotational speed (rotor angular speed ω) of the electric motor 7.
Figure 10 is a configuration in which do not change the carrier frequency f _c in the PWM converter section 4, FIG. 11 is a configuration for changing the carrier frequency f _c.

10 and 11, t _u , t _v and t _w are pulse widths of u, v and w phases before dead time compensation, and t _u ′, t _v ′ and t _w ′ are after dead time compensation. The pulse width of each phase of u, v, and w. Moreover, 17-19, 17'-19 ', 20, 20', and 21-23 are the same as the thing of the same code | symbol in FIG. 6, FIG. Reference numeral 27 denotes a switch, which can be switched to a side where the gain of the variable gain multiplier is set to 1 in the stop state or in the low speed rotation according to the rotation speed (rotor angular speed ω) of the electric motor 7, and in the middle or high speed rotation. It is switched to the map 20, 20 ′ side.
Regarding the rotation speed for switching the switch 27, the configuration using only the three-phase current (the side where the gain of the variable gain multiplier is set to 1) is difficult to adjust during medium / high speed rotation, and has an error in control. The rotational speed exceeding the possible error range is determined by experiment.

Also, if there is a rotational speed dependent on the correction value by the magnitude i _a of the current vector may be a map 20, 20 'as a value for each rotational speed change.
Each map may be created by a method of setting a compensation amount that matches the voltage command value and the actual voltage, or a method for preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched. (It is not equal to the actual voltage). These maps may be created using values obtained by experiments in advance, or approximate mathematical formulas may be used.

  The method of setting the compensation amount to match the voltage command value and the actual voltage has an effect that stable operation can be performed because no error is added to the constant of the motor 7 to be controlled. In addition, the method of preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched has the effect of simplifying the configuration and reducing the memory capacity and calculation load in a system that switches the carrier frequency. is there.

Further, in the case of setting the compensation amount so that the voltage command value and the actual voltage coincide with each other, the configuration shown in FIG. 10 may be used unless the carrier frequency is changed, and the configuration shown in FIG. 11 is used when the carrier frequency is changed.
In the method of preventing the relationship between the voltage command value and the actual voltage from changing even when the carrier frequency is switched, the configuration of FIG. 10 may be used even when the carrier frequency is changed.
As described above, the magnitude of the current vector, the u-phase current value, the v-phase current value, and the w-phase current value may be an actual current value, a current target value, or a current estimated value estimated from the current target value. A combination of these may be used, but operation is more stable when the actual current value is not used.

When the current control operates stably, the current estimated value is operated more stably than the current target value is used.
In addition, reference maps may be the torque command value T ^* instead of i _a, may be torque estimated value T ^ multiplied by the filter to the torque command value T ^*.

  As described above, the adjustment using the three-phase current is difficult to adjust during the middle / high-speed rotation. However, in the dead time compensation example 4, the adjustment becomes simple and the dead time compensation with high accuracy is performed in the entire operation region. be able to.

Here, the reason why the dead time compensation is necessary in the motor control apparatus to which the present invention is applied will be described.
FIG. 14 is a diagram showing a circuit and a switching waveform showing one phase of the switching elements constituting the inverter (example of u phase).
In FIG. 14, Vdc is a power supply voltage, Vu is a u-phase voltage, iu is a u-phase current, UP is a drive signal for a positive-side switching element, UN is a drive signal for a negative-side switching element, and τ is a dead time.
As shown in the figure, an inverter that drives an electric motor is configured by a circuit in which two switching elements (power MOSFET, IGBT, etc.) are cascade-connected (cascade connection) for each phase.

In operation, one of the two switching elements is turned on and the other is turned off, and by switching this, a positive (u-phase current iu> 0) or negative (u-phase current iu <0) current is applied to the motor winding. It is supposed to flow. In order to prevent the two switching elements from being turned on at the same time, control is performed so that the other is turned on after a predetermined dead time τ after one of the switching elements is turned off.
In the circuit as described above, as shown in the waveform diagram, since the on-time of the negative side switching element is longer by the dead time τ, the switching element itself has exactly the same operation waveform, but the current The waveform of the actual voltage applied (the waveform at the bottom in the figure) differs depending on the positive / negative of. Preventing this is dead time compensation, and it is fundamental to determine the direction of current and change the PWM signal for driving the switching element (correct the pulse width).

1 is a block diagram showing the configuration of Embodiment 1 of the present invention. The block diagram which shows the structure of Example 2 of this invention. The block diagram which shows the structure of Example 3 of this invention. FIG. 3 is a block diagram illustrating a first embodiment of a dead time compensation unit 5; FIG. 3 is a block diagram illustrating a first embodiment of a dead time compensation unit 5; FIG. 6 is a block diagram illustrating a second embodiment of a dead time compensation unit 5; FIG. 6 is a block diagram illustrating a second embodiment of a dead time compensation unit 5; FIG. 6 is a block diagram illustrating a third embodiment of the dead time compensation unit 5; FIG. 6 is a block diagram illustrating a third embodiment of the dead time compensation unit 5; FIG. 9 is a block diagram illustrating a fourth embodiment of the dead time compensation unit 5. FIG. 9 is a block diagram illustrating a fourth embodiment of the dead time compensation unit 5. The figure which shows an example of the approximate characteristic of a compensation amount. The figure which shows another example of the approximate characteristic of a compensation amount. The figure which shows the circuit and switching waveform for 1 phase of the switching element which comprises an inverter.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Current map part 2 ... Current control part 3 ... Three phase conversion part 4 ... PWM conversion part 5 ... Dead time compensation part 6 ... Inverter 7 ... Electric motor 8 ... dq axis conversion part 9 ... Differentiation part 10 ... Equivalent to current control Filter 11, 11 '... Three-phase conversion unit 12, 13 ... Compensation amount map 14, 15, 16 ... Polarity setting unit 17-19 ... Compensation amount map 17'-19' ... Compensation amount map 20, 20 '... Correction amount map 21-23 ... Variable gain multipliers 24-26 ... Switch 27 ... Switch

Claims (7)

The dq-axis voltage command value obtained from the deviation between the dq-axis current target value based on the torque command value and the actual current value is three-phase converted into a three-phase voltage command value. In the motor control device that drives the motor by supplying three-phase AC power by controlling the inverter with the PWM signal that has been PWM-converted,
Between the PWM conversion means for performing the PWM conversion and the inverter, a dead time compensation means for compensating for a dead time of a switching element constituting the inverter is provided,
The dead time compensation means is:
Based on the current vector magnitude ia obtained from the d-axis current i _d and the q-axis current i _q by the formula i _a = √ ( _id ² + i _q ² ), the magnitude of the dead time compensation amount is set,
A motor control apparatus, wherein the polarity of the compensation amount is switched according to a three-phase current value flowing in each phase of the motor. The dq-axis voltage command value obtained from the deviation between the dq-axis current target value based on the torque command value and the actual current value is three-phase converted into a three-phase voltage command value. In the motor control device that drives the motor by supplying three-phase AC power by controlling the inverter with the PWM signal that has been PWM-converted,
Between the PWM conversion means for performing the PWM conversion and the inverter, a dead time compensation means for compensating for a dead time of a switching element constituting the inverter is provided,
The dead time compensation means is:
A dead time compensation amount is set based on the magnitude ia of the current vector obtained from the d-axis current i _d and the q-axis current i _q by the formula i _a = √ ( _id ² + i _q ² ),
Set the basic compensation amount based on the three-phase current flowing through each phase of the motor, and compensation amount correction value based on the magnitude i _a of the current vector compensation amount that is the set <br / > An electric motor control device characterized by the above. The dq-axis voltage command value obtained from the deviation between the dq-axis current target value based on the torque command value and the actual current value is three-phase converted into a three-phase voltage command value. In the motor control device that drives the motor by supplying three-phase AC power by controlling the inverter with the PWM signal that has been PWM-converted,
Between the PWM conversion means for performing the PWM conversion and the inverter, a dead time compensation means for compensating for a dead time of a switching element constituting the inverter is provided,
The dead time compensation means is:
A dead time compensation amount is set based on the current vector magnitude ia obtained from the d-axis current i _d and the q-axis current i _q by the formula i _a = √ ( _id ² + i _q ² ).
In a stopped state or a low-speed rotation area of the motor, to set the compensation amount based on the three-phase current flowing through each phase of the motor, in the high-speed rotation region in the electric motor on the basis of the size i _a of the current vector compensating A control device for an electric motor characterized by switching to set an amount . The dead time compensation means sets a basic compensation amount based on the three-phase current flowing through each phase of the motor, a value corrected based on the compensation amount and the set to the size i _a of the current vector and the compensation amount, control of the electric motor according to claim 2, wherein the switching the correction amount when correcting the magnitude i _a of the current vector in said electric motor in a stopped state or a low-speed rotation region and the middle speed rotation region apparatus. 5. The motor control device according to claim 1 , wherein the compensation amount is set to a value that matches the three-phase voltage command value and the actual voltage output from the inverter. 6. The relationship between the size i _a and the compensation amount of the current vector, is stored by approximating the memory by a simple linear expression by a combination of straight lines, on the basis of the size i _a of the linear expression and the current vector 4. The motor control device according to claim 1 , wherein a compensation amount is set. As the size i _a prior SL current vector, the current estimated value or the d-q axis current target value of the d-q-axis estimated from the d-q axis current target value d-q axis current values id, as iq the motor controller according to any one of formulas _{i a = √ (i d 2} + i q 2) according to claim 1 to claim 6, characterized by using a value obtained by using.

Images