JPH0772163A - Detection apparatus for rotational speed of ac motor - Google Patents

Abstract

PURPOSE:To detect the rotational speed without using a speed sensor by a method wherein a voltage detection means, a current detection means and a rotational-speed operation means are provided. CONSTITUTION:In order to lower the cost of the apparatus, a speed sensor is not installed. A rotational-speed operation and magnetic-flux control circuit 50 is installed so that a slip angular velocity omegas is operated on the basis of the output current and the output voltage of an inverter 2 and that a magnetic flux detected. Consequently, output lines 31a which output the output-voltage instantaneous value of the inverter 2 and output lines 34a which detect the input-current instantaneous value of a motor 1 are connected to the circuit 50. A motor rotational speed omegam which is found by the circuit 50 is input to a subtracter 10, it is compared with the desired rotational speed (speed reference signal) of a line 11, and a difference signal between both is obtained. The output of the subtracter 10 is input to a part 12, and a difference signal VD is obtained in an output line 13. A generated triagular-wave voltage VC is compared 16 with the absolute value of the difference signal VD by a comparator 16, and a drive or a stop is decided. The comparator 16 is connected to an AND gate 18 and a ROM 5. The inverter 2 is controlled so as to correspond to the voltage vector data of the ROM 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for detecting the rotation speed of an AC motor driven by a PWM inverter without using a speed sensor.

[0002]

2. Description of the Related Art As a speed estimation method for a speed sensorless drive system, one using the property of vector control,
It is known that a model reference adaptation system is used. When vector control is achieved in the former method, the torque and torque current (torque component on the primary side of the motor) are proportional, so the slip angular velocity is calculated using the torque current and secondary magnetic flux to rotate the induction motor. The speed can be estimated. In the latter model reference adaptive system, the same input as the actual system is given to the model using the characteristic equation of the induction motor, and the error of those outputs is assumed to be due to the error of the speed ω _m. Estimate the velocity from the output error.

[0003]

A secondary resistance that cannot be directly measured is used in the slip angular velocity calculation formulas and model characteristic equations of these methods, and the measured value of the secondary resistance and the actual system are used. If the values of do not match, this leads to an error in speed estimation. Further, since the secondary resistance value changes along with the temperature rise and the skin effect at the time of transition, the speed estimation characteristic deteriorates even during operation in the conventional method.

SUMMARY OF THE INVENTION An object of the present invention is to provide an apparatus capable of detecting the rotational speed of an AC motor relatively easily and moment by moment.

[0005]

SUMMARY OF THE INVENTION To achieve the above object, the present invention is a device for detecting the rotation speed of a three-phase AC motor driven by a three-phase PWM inverter, wherein the three-phase PWM inverter is used. Three-phase output voltage instantaneous value (v _1a , v _1b ,
v _1c ), voltage detection means for detecting the three-phase output current instantaneous value (i _1a , i _1b , i _1c ) of the three-phase PWM inverter, and three-phase output voltage instantaneous value (v _1a , v _1b , v _1c ) are two-phase voltage instantaneous values (v _1d , v _1q ).
And the three-phase output current instantaneous values (i _1a , i _1b ,
i _1c ) is converted into a two-phase current instantaneous value (i _1d , i _1q ), and the two-phase output voltage instantaneous value (v _1d , v _1q ) and the two-phase output current instantaneous value (i _1d , i _1q ) are converted into The primary magnetic flux (φ _1d , φ _1q ) and the secondary magnetic flux (φ _2d , φ _2q ) of the AC motor are obtained based on the above equation, and the instantaneous values of the two-phase output current (i _1d , i _1q ) and the temporary magnetic flux are obtained. (Φ _1d , φ _1q ), the secondary magnetic flux (φ _2d , φ _2q ), and the primary, secondary and mutual inductances (L) of the AC motor.
Those related to the rotational speed detecting device of the motor - AC motor, characterized in that it consists of a rotational speed calculating means for calculating a value corresponding to the rotational speed of the motor - _11, wherein the AC motor based on the L _22, M) .

[0006]

In the present invention, the rotation speed of the motor is estimated without using the secondary resistance. Therefore, it becomes possible to detect the rotation speed regardless of the change in the secondary resistance value due to the temperature change or the like. Since the rotation speed can be known without using the speed sensor, the cost and size of the device can be reduced.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, a rotation speed control device for a three-phase AC motor (induction motor) using an impulsive torque drive inverter device according to an embodiment of the present invention will be described. In FIG. 1 showing this speed control device, a PWM-controllable three-phase inverter 2 is connected to a motor 1 composed of a three-phase induction motor. The inverter 2 includes switching elements Q1, Q2, Q3, Q4, which are transistors in the DC power supply 3,
It is a bridge connection of Q5 and Q6. The six switch elements Q1 to Q6 are turned on / off in response to a control signal supplied from the drive circuit 4. Note that the upper three switch elements Q1, Q2, Q3 and the lower three switch elements Q4, Q5, Q6 of the inverter 2 operate in reverse to each other, so if one control is specified, the entire inverter Controls are identified. Here, the inverter control state is specified by the first, second, and third signals A, B, and C read from the ROM (Read Only Memory) 5, and the signals A, B, and C are set to the high level, that is, the logic "1". The switch elements Q1, Q2, Q3 are turned on when "", and the switch elements Q1, Q2, Q3 are turned off when they are at a low level, that is, a logical "0".

[0008]

[Description of ROM address] ROM5 is the inverter 2 PW
A PWM switching pattern (unit vector data) for M control is written in advance. This RO
M5 is a forward rotation PWM pattern memory M1, M5, a forward rotation zero vector memory M2, a reverse rotation PWM pattern memory M3, M7, a reverse rotation zero vector memory M4, a forward rotation normal vector M6, and a reverse rotation normal vector M8. Have. Each of the memories M1 to M8 is, for example, one of 0 to 1023
024 addresses, each addressed by the value of the 10-bit binary output line 6a of the up / down counter 6. However, one is selected from the eight memories M1 to M8, and only the output of the selected memory is effectively used for controlling the inverter 2. In order to make this selection, the ROM 5 has a zero vector selection control terminal 7, a forward / reverse rotation selection control signal input terminal 8, and a normal vector selection control terminal 42. First, when the normal vector selection control signal terminal 42 is logic "0", M1 to M4 including the zero vector memory are selected. When the logical value is "1", M5 to M8 including the normal vector memory are selected. When the zero vector selection control signal input terminal 7 is logic "0", either one of the memories M1 and M3 or M5 and M7 is selected. When the zero vector selection control signal input terminal 7 is logic "1", the memories M2 and M4 or M6 and M8 are selected. Either one is selected. Furthermore,
When the forward / reverse rotation selection control signal input terminal 8 is "0", one of the memories M1 and M2 or M5 and M6 is selected, and when it is "1", the memories M3 and M4 or M.
Either one of 7 and M8 is selected. Now line 6
If the 10 bits of a are represented by 10 bits of A0 to A9, the input bit of the input terminal 7 is represented by A10, the input bit of the input terminal 8 is represented by A11, and the input bit of the input terminal 42 is represented by A12, then An address is designated by 10 bits of A0 to A9. In addition, A10, A11, A12 are [A12,
Expressed as A11, A10], the first memory M1 (normal rotation PWM switching pattern) is selected at [000] and the second memory M2 (normal rotation zero vector) is selected at [001]. In [111], M8 (reverse normal vector) is selected.

In order to control the rotation speed of the motor 1, information on the rotation speed is required. Conventionally, the rotational speed is detected by the speed detector, but in the present embodiment, the speed estimating means is provided without providing the speed detector in order to reduce the cost of the device. That is, there is provided a rotation speed calculation / flux control circuit 50 for estimating the rotation speed of the motor 1 based on the output current and output voltage of the inverter 2 and controlling the magnetic flux. Details of the rotation speed estimation method will be described later.

The motor rotation speed ω _m obtained by the rotation speed calculation / flux control circuit 50 is input from a line 9a to a subtracter 10 as a difference signal forming means (comparing means), and a line 11 as a speed command generating means. Of the digital speed command (desired rotation speed), that is, a reference signal, and a difference signal between the two is obtained from the subtracter 10.

The output of the subtractor 10 _is input to a proportional-plus-integral compensation circuit 12 represented by K (1 + 1 / T _is ), and a compensation output, that is, a difference signal V _D is obtained at this output line 13. This compensation output indicates the operation amount in control, and the reading of vector data from the memory 5 is determined based on this.

The signal compensation difference signal V _{D of the} compensation output line 13
Is input to the first comparator 14 for determining whether the difference signal V _D is positive or negative, and the second comparator 14 passes through the absolute value circuit 15.
Input to the comparator 16. The output terminal of the first comparator 14 for determining the forward / reverse rotation (F / B) is connected to the up / down input terminal U / D of the counter 6 and the forward / reverse rotation selection signal input terminal of the ROM 5 is connected. 8 is connected.

Reference numeral 17 denotes an oscillator (OSC), which has 20 to 20
A clock pulse of about 50kHz is generated. The output terminal of the oscillator 17 is connected to one input terminal of the AND gate 18, and the output terminal of the AND gate 18 is connected to the clock input terminal CL of the counter 6, so that AN
The output of the oscillator 17 is input to the counter 6 as a clock pulse only when the other input terminal of the D gate 18 is at high level.

A triangular wave generator 19 is connected to the non-inverting input terminal of the second comparator 16 for determining drive / stop. The triangular wave generator 19 generates a triangular wave voltage Vc (carrier) at 2.5 kHz, which is lower than the output frequency of the oscillator 17, and the comparator 16 compares this Vc with the absolute value of the difference signal V _D. The output terminal of the second comparator 16 is N
It is connected to the input terminal of the AND gate 18 via the OT circuit 20 and also connected to the zero vector selection control signal input terminal 7 of the ROM 5.

The rotation speed calculation and magnetic flux control circuit 50 determines the slip angular speed ω _s based on the output voltage and current of the inverter 2.
Is calculated and the magnetic flux is detected. Therefore, the three output lines 31a for detecting the instantaneous value of the output voltage of the inverter 2 and the three output lines 34a of the current sensors 43a, 43b, 43c for detecting the instantaneous value of the input current of the motor 1 are the rotational speeds. The output line 9a of this rotation speed ω _m is connected to the arithmetic and magnetic flux control circuit 50.
Is connected to the subtractor 10 and the magnetic flux signal output line 55 is connected to the terminal 42 of the ROM 5. Although the circuit for controlling the reading of the vector data (switching pattern) from the memory 5 is shown in FIG. 1 in an analog manner, actually, the subtractor 10, the proportional-plus-integral compensation circuit 12, the comparator 14, 16, the absolute value circuit 15, the triangular wave generation circuit 19, the rotation speed calculation and magnetic flux control circuit 50 are DS
P (digital signal processing device). Further, an A / D converter is connected to the output voltage detection line 31a and the current detection line 34a of the inverter 2, but it is omitted in FIG. 1 to simplify the drawing.

FIG. 2 shows the rotation speed calculation and magnetic flux control circuit 50 of FIG. 1 in detail. Before describing each part of the circuit 50, the principle of the method for detecting the primary interlinkage magnetic flux will be described.

The characteristic equation of the induction motor can be expressed by the following equation (1).

[0018]

[Equation 1]

Here, v _1d and v _1q are two-phase primary voltages of the motor 1 whose primary voltages v _1a , v _1b and v _1c are shown by dq coordinate axes, and i _1d and i _1q are primary currents i of the motor 1. _1a , i _1b , i _1c
Are primary and secondary winding resistances, L ₁₁ and L ₁₂ are primary and secondary winding inductances, and ω _m is a rotor rotation angular velocity (rotational speed). ,
M is the mutual inductance of the primary and secondary windings, and p is d / dt
Is a differential operator.

The undetectable values in the equation (1) are the secondary currents i _2d and i _2q . The secondary currents i _2d and i _2q are obtained by the first and second rows of the equation (1). Substituting these into the third and fourth rows of the equation (1), the rotational angular velocity ω _m
Two equations (2) and (3) for

[0021]

[Equation 2]

By the way, since the undetectable secondary resistance R2 is included in both the equations (2) and (3), the equation (2)
And the equation (3) are combined to eliminate the secondary resistance R2 to obtain the following equation (4).

[0023]

[Equation 3]

Here, φ _1d and φ _1q are two-phase primary interlinkage magnetic fluxes showing the three-phase primary interlinkage magnetic flux in the motor 1 on the dq coordinate axes, and φ _2d and φ _2q are three-phase primary interlinkage fluxes in the motor 2. It is a two-phase secondary interlinkage magnetic flux that shows the next interlinkage magnetic flux on the dq coordinate axes.

The primary interlinkage magnetic flux vector φ ₁
And the primary voltage vector v ₁ and the relationship between the primary current vector i ₁ and the primary resistance R ₁ are given by the following equation (5), and the secondary interlinkage magnetic flux vector φ ₂ and the primary interlinkage magnetic flux vector φ ₁
The following equation (6) shows the relationship between the primary and secondary inductances L ₁ and L ₂ and the mutual inductance M.

[0026]

[Equation 4]

Φ _1d and φ _1q are expressed by the following equation (7) in the same manner as φ _1.
And φ _2d and φ _2q can be expressed by the following equation (8) similarly to φ ₂ .

[0028]

[Equation 5]

V _1d in the equations (7) and (8),
v _1q , R ₁ , i _1d and i _1q are values that can be determined based on detection or measurement or calculation. Further, the parameters of the equations (9) and (10) are also values that can be determined based on detection, measurement, or calculation. Therefore, the rotational angular velocity ω _{m in} equation (4) can be determined by calculation based on the detected value.

The two-phase output voltages v _1d and v _1q and the current i _1d indicated by the dq coordinate axes included in the equation (1)
i _1q is the instantaneous value of the three-phase output voltage of the inverter 2 (primary voltage)
v _1a , v _1b , v _1c and three-phase output current instantaneous value (primary current)
Based on i _1a , i _1b , and i _1c , the following formula (11) and formula (1
Determined in 2).

[0031]

[Equation 5]

Next, the rotation speed calculation / flux control circuit 50 of FIG. 1 will be described in detail with reference to FIGS. 2, 3 and 4.
FIG. 2 shows a functional block diagram of the rotation speed calculation / flux control circuit 50, that is, an equivalent circuit. In FIG. 2, the output line 31a of the inverter 2 is connected to the three-phase / two-phase conversion circuit 31 including an operational amplifier. In the three-phase / two-phase conversion circuit 31, the output voltage v _{1a of} the inverter 2 is calculated according to the equation (11),
The v _1b and v _1c are converted into two-phase output voltages v _1d and v _1q indicated by the dq coordinate axes. The current detection line 34a of the inverter 2 is connected to the three-phase / two-phase conversion circuit 34 including an operational amplifier. The conversion circuit 34 converts the inverter output currents i _1a , i _1b , i _1c into two-phase currents i _1d , i _1q indicated by the dq coordinate axes according to the equation (12).

The φ _1d and φ _1q arithmetic circuits 60 are connected to the voltage and current three-phase / two-phase conversion circuits 31 and 34, respectively,
Based on v _1d , v _1q , i _1d , and i _1q , the operations of equations (7) and (8) are executed, and the primary magnetic fluxes φ _1d and φ _1q expressed in Cartesian coordinates.
Is output. As shown in FIG.
Two multipliers 61 and 6 for multiplying _1d and i _1q by R _1.
2 and two subtractors 63 and 64 for subtracting v _1d and v _1q from i _1d R ₁ and i _1q R ₁ and two integrators 65 and 66 for integrating these outputs. be able to.

As shown in FIG. 2, φ _2d , φ _2q arithmetic circuit 70
Is a φ _1d , φ _1q arithmetic circuit 60 and a current three-phase / two-phase converter 3
The secondary magnetic fluxes φ _2d and φ _2q of the motor 1 are calculated. The details of the φ _2d and φ _2q arithmetic circuit 70 include two L ₂₂ / M multipliers 71 and 72 and two (L ₁₁ L ₂₂ −M ² ) / M multipliers 73 and 74 as shown in FIG. It is composed of two subtractors 75 and 76, and executes the operations of equations (9) and (10).

The rotational angular velocity calculating circuit 80 in FIG. 2, phi _1d,
and phi _1q arithmetic circuit 60, φ _2d, φ _2q is connected to an operation circuit 70 and the current three-phase two-phase conversion circuit 34, Equation (4) to calculate the L ₁₁ and L ₂₂ as shown in FIG. 4 Multipliers (coefficient multipliers) 91 and 92, two subtractors 93 and 94, two differentiators 95 and 96, and four multipliers 97, 98 and 99,
100, a subtractor 101 and an adder 10 at the output stage of these
2 and one divider 103. Considering that the motor is driven by a sinusoidal voltage, the numerator and denominator of equation (4) are both zero in the steady state, and this equation cannot be used. However, here, since the motor 1 is driven by the PWM inverter, the motor 1 is driven by the waveform in which the high frequency ripple is superimposed, the steady state disappears due to the effect of the superimposed high frequency ripple, and the numerator and denominator are not zero. Instead, it becomes possible to obtain the rotational angular velocity ω _m by the equation (4).

Next, it will be described that the numerator and denominator of the equation (4) become zero by the sine wave drive and that they do not become zero by the PWM control. The expression (4) can be expressed by the following expression (13) for the three phases collectively. ω _m = | (φ ₁ −L ₁ i ₁ ) × (d / dt) φ ₂ | / (φ ₁ −L ₁ i ₁ ) · φ ₂ = | Mi ₂ × (d / dt) φ ₂ | / Mi ₂ · φ ₂ (13) where i 2 is the secondary current vector. When the motor 1 is a sine wave drive and the slip S is constant, d / dtφ
₂ and i ₂ are in phase, and φ ₂ and i ₂ are orthogonal. Therefore, in the numerator of the equation (13), since d / dtφ ₂ and i ₂ are in phase, the magnitude of the outer product is | d / dtφ ₂ || i ₂ | sin 0 = 0. Since φ ₂ and i ₂ are orthogonal to each other in the denominator of Expression (13), the value of the inner product is | φ ₂ || i ₂ | cos π / 2 = 0. On the other hand, when a high frequency ripple is superimposed on the drive voltage of the motor 1 based on PWM control, the rotation of the secondary magnetic flux repeats normal rotation and reverse rotation instantaneously, and the high frequency components of φ ₂ and i ₂ are no longer orthogonal. .

FIG. 10 shows a calculated value waveform of the rotational angular velocity ω _m according to the equation (1). As shown in (a) of FIG. 10, the speed calculation value can be stably obtained. On the other hand, the values of the numerator and the denominator of the equation (1) fluctuate as shown in (b) and (c) of FIG. As shown in (c) of FIG. 10, a period in which the denominator becomes zero instantaneously occurs. Therefore, a velocity calculation error occurs at or near the place where the denominator becomes zero. In order to eliminate the influence of this error, the calculation is not performed near the denominator becomes zero, and L
Speed control equivalent to that smoothed by PF is executed.

FIG. 11 shows the actual speed of the motor 1 in the apparatus of FIG. 1, the calculated speed, the load torque, and the calculated value of the secondary resistance R2. In FIG. 11, the motor 1 is moved in the first direction by 180
After rotating at 0 rpm, rotate at 1800 rpm in the second direction, keep steady state, and then restart at 18 rpm in the first direction.
Although it is rotating at 00 rpm, the calculation speed corresponds well to this. Further, the calculated speed is substantially equal to the actual speed even in the period in which the load is in the rated load state and the load torque is generated. Further, the fluctuation of the secondary resistance R2 does not affect the calculation speed.

The φ ₁ arithmetic circuit 39 of FIG. 2 is connected to the φ _1d and φ _1q arithmetic circuits 60, and φ ₁ = (φ _1d ² + φ _1q ² ) ^1/2
Is executed and the absolute value of the primary magnetic flux φ ₁ is output. Δ
Hysteresis comparators 40 and 41 having a hysteresis comparison width of | φ ₁ | compare the detected value | φ ₁ | of the primary interlinkage magnetic flux with the primary interlinkage magnetic flux command value | φ _1a |. And | φ
_{When 1} | further exceeds | φ _1a | + Δ | φ ₁ | and the comparator 41 outputs a logic "0", the blocks M1 to M4 containing the zero vector in the ROM 5 are selected. Also, |
When φ ₁ | exceeds | φ _1a | −Δ | φ ₁ | and further decreases, a logic “1” is output and the blocks M 5 to M 8 including the normal vector are selected in the ROM 5. As a result, the magnitude of the primary magnetic flux φ ₁ is controlled to be constant.

Controlling an inverter based on a voltage vector and a zero vector describes an impulse torque drive.

[Contents of ROM] Data is written in the ROM 5 as shown in principle. That is, the ROM has addresses 0 to 1023, but in FIG.
The arrangement of vectors for an address of 1 is shown. Forward PW
For example, data of voltage vectors V6, V2, V6 and V2 are sequentially written in the addresses 0 to 3 of the M pattern memories M1 and M5, and zero vectors V7, V0 and V7 are written in the addresses 0 to 3 of the zero vector memory M2 for normal rotation. , V0 data are written in order, and the reverse PWM pattern memories M3, M7 are written.
At addresses 0 to 3 of voltage vectors V1, V5, V1
, V5 are sequentially written, and the zero vector memory M4 for reverse rotation is sequentially written with the data of zero vectors V0, V7, V0, V7, and the normal vector memory M for normal rotation is written.
Forward rotation PWM pattern memory M is assigned to addresses 0 to 3 of 6.
The normal vector V4 corresponding to the vector of addresses 0 to 3 of 1 is written, and the normals corresponding to the vectors of addresses 0 to 3 of the PWM pattern memory for reverse M3 are written in the addresses 0 to 3 of the normal vector memory for reverse M8. Vector V4 is written. Vector data is also written in the remaining addresses 4 to 511 according to the same principle as the addresses 0 to 3. The vector data of each address in FIG. 5 shows the principle and is different from the actual data. Now, the normal rotation PWM pattern memory M1 addresses 0 to 84 (0 degrees to
Actual voltage vector data (corresponding to 60 degree section) is shown as V6, V6, V6, V6, V2, V2, V2, V2.
, V2, V2, V6, V6, V6, V6, V2, V2
, V2, V2, V2, V2, V6, V6, V6, V6
, V2, V2, V2, V2, V2, V2, V2, V2
, V2, V2, V2, V2, V2, V2, V2, V2
, V2, V2, V2, V2, V2, V2, V2, V2
, V2, V2, V2, V2, V2, V2, V2, V2
, V2, V2, V2, V2, V2, V3, V3, V3
, V3, V2, V2, V2, V2, V2, V2, V3
, V3, V3, V3, V2, V2, V2, V2, V2
, V2, V3, V3, V3, V3.

FIG. 6 shows six voltage vectors V1 to V6,
Two zero vectors V0 and V7 are shown. The switching states of the switching elements Q1, Q2, Q3 of the inverter 2 are (000), (001), (010), (01
1), (100), (101), (110), (11
Since there are eight of 1), this is V0, V1, V2, V
3, V4, V5, V6 and V7. In the apparatus of this embodiment, the voltage vectors V0 to V7 are written in the ROM 5 and output as control data (A, B, C). Combining the eight vectors V0 to V7,
A sinusoidal output voltage and a rotating magnetic field vector can be obtained.

[0042]

[Vector Selection] FIG. 7 shows the selection of the voltage vector for obtaining the rotating magnetic field vector φ ₁ . In order to bring the locus of the tip (end point) of the rotating magnetic field vector φ ₁ close to a circle, the sixth and second vectors V in the 330 ° to 30 ° section.
6, V2, second and third vectors V2, V3 in the 30-90 degree section, third and first vectors V3, V1, 90-150 degree section, the first and 150-210 degree sections, and Fifth and fourth vectors V5, V4, 270-33 in the fifth vector V1, V5, 210-270 degree section.
The fourth and sixth vectors V4 and V6 are selected in the 0 degree section. In principle, V6 and V2 are selected as the significant vectors in the 330 to 30 degree section of FIG. 7, and the zero vector V7 is selected when the vector rotation is stopped. Also,
The normal vector is a vector extending in the radial direction from the center of the circular locus of the magnetic flux, and V4 is selected in the section of 330 to 30 degrees and V6 is selected in the section of 30 to 90 degrees in FIG. When the motor 1 is normally rotated, the rotating magnetic field vector .phi.1 is rotated in the direction indicated by UP in FIG. 7, and when the motor 1 is rotated in the reverse direction or is braked, it is rotated in the direction indicated by DOWN.

[0043]

[Operation] Next, the braking operation of the circuit of FIG. 1 will be described with reference to FIGS. When the difference signal VD is obtained on the basis of the comparison between the estimated speed signal obtained on the line 9a and the reference signal (target signal) on the line 11, the positive / negative of this signal is judged by the first comparator 14, and now the positive signal If it is a signal, FIG.
The comparison output becomes a low level "0" as shown before t4 in (C), and this is input to the counter 6. Therefore, the counter 6 operates up during this period. In the second comparator 16, the absolute value of the difference signal V _D and the triangular wave voltage VC are compared as shown in FIG. 8 (A), and the output of FIG. 8 (B) is generated. In other words, generates a high level output "1" when the triangular wave voltage VC is higher than the absolute value of the difference signal V _D (t1 ~t2), a low-level output "0" when the opposite (t2 -t3) Occur. As in t1 to t2, the output bit A10 of the second comparator 16 is at high level "1", the output bit A11 of the first comparator 14 is at low level "0", and further, in FIG. As shown in D), t10 when the outputs of the hysteresis comparators 40 and 41 in FIG. 3 are low level (L).
In the previous time, the ROM 5 [A12 A11 A1
0] = [001], the forward zero vector memory M2 is selected, and [A12 A11 A
When 10] = [000], the normal PWM pattern M1 is selected. Further, during a period (t1 to t2) in which the output of the second comparator 16 is at a high level (H), the output of the NOT circuit 20 becomes a low level, and the clock pulse of the oscillator 17 may pass through the AND gate 18. Blocked, counter 6
Is not incremented, so keep specifying the same address. On the other hand, during the period when the output of the second comparator 16 is low level (t2 to t3), the output of the NOT circuit 20 becomes high level, so the output clock pulse of the oscillator 17 is ANDed.
It passes through the gate 18 and becomes an input pulse of the counter 6.
As a result, the values of 10 bits A0 to A9 of the counter 6 are increased by the up operation, and the addresses of the memory M1 are sequentially designated. However, if the output of the second comparator 16 goes high at time t3, the clock input of the counter 6 is prohibited and the counter 6 retains the addressing at this time. For example, when the memory M2 is selected while the vector V6 of the memory M1 is being read at the address 2 as shown in FIG. 5, the forward zero vector V7 (111) at the same address 2 is selected. The zero vector V7 is the second
The output of the comparator 16 continues to be generated during the high level, the comparison output returns to the low level, the clock pulse of the counter 6 is input again, and when the output of the counter 6 is incremented by one stage, the normal rotation PWM pattern memory The voltage vector V2 (010) of address 3 of M1 is selected. The zero vector consists of two types, V0 (000) and V7 (111), and the vector that requires less switching of the switch elements Q1 to Q6 is selected. Counter 6 is a decimal number 0
When the binary number corresponding to 1023 has been generated, the normal rotation PWM
All the voltage vector data of 0 to 360 degrees of the pattern is read out, the three-phase approximate sine wave voltage is generated from the inverter 2, and the rotating magnetic field vector close to the circular locus is generated in the motor 1.

In such control, when the difference between the target rotation speed and the estimated speed becomes small, the period in which the output of the second comparator 16 becomes high level becomes relatively long and the zero vector is selected. Becomes longer.

If the level of the reference signal on the line 11 is lowered to the low speed rotation command state from t20 to t4, the level of the absolute value of the difference signal V _D is also lowered and the output frequency f of the inverter 2 is lowered. At the same time, the output voltage V also drops, and the motor 1 is driven at a low speed.

At t4 in FIG. 8, switching to the reverse rotation command,
When the difference signal V _D becomes negative, the output of the first comparator 14 becomes high level and reverse control is performed. In the PWM control, when the voltage vector is switched,
Since a pair of switch elements Q1, Q4, or Q2, Q5, or Q3, Q6 may be short-circuited by storage etc. and they may be destroyed, an uncontrolled period is provided between the vectors to prevent this. Is desirable.

The period from t10 to t20 in FIG. 8 is a period in which the outputs of the hysteresis comparators 40 and 41 in FIG. 3 are at a high level. That is, when the speed reference signal on the line 11 in FIG. 3 sharply increases at t10, the difference signal V _D sharply increases as shown in FIG. 8A, and thus the period for outputting the normal rotation vector sharply increases. It operates to increase the rotation speed of the motor rapidly. As a result, the primary current of the motor rapidly increases to obtain the desired acceleration. However, the voltage drop due to the primary winding resistance of the motor also increases, and the primary flux linkage of the motor | φ
On the contrary, ₁ | decreases as after t10 in FIG.
| Φ _1a | -Δ | φ ₁ | When the voltage drops to or below, the hysteresis comparator operates and outputs a high level. At this time, ROM
5, the terminal 42 becomes high level, so that the blocks M5 to M8 containing the normal vector are selected. Therefore, t
In the period from 10 to t20 and the second comparator is in the high level period t
In 11 to t12, the normal vector for normal rotation M6 is selected in place of the zero vector memory for normal rotation M2. FIG. 9 shows 30
This state is shown in the interval of 90 degrees to 90 degrees.
Similarly, the second comparator is in the low level period t12 to t13.
Is selected as M5. Thus the primary interlinkage magnetic flux of the motor by outputting a normal vector in place of the zero vector of size is increased motor primary interlinkage magnetic flux | phi ₁ | a _{| φ 1a | -Δ | φ 1} | more It becomes possible to As a result, the motor can obtain a desired acceleration, follow the increase of the speed reference signal with good response, and reach a desired rotation speed. Then, for example, when the level of the speed reference signal on line 11 decreases and the primary current decreases, the voltage drop due to the primary winding resistance also decreases and the primary interlinkage flux consequently increases. At t20, the primary flux linkage is | φ _1a | + Δ | φ ₁ |
The output of the hysteresis comparator goes to a low level, and therefore the terminal 42 goes to a low level in the ROM 5, so that the blocks M1 to M4 containing the zero vector are selected and the primary interlinkage magnetic flux is reduced to | φ _1a | + Δ | φ ₁ | From the above, the primary flux linkage of the motor | φ ₁ |
It will be controlled within the range of φ _1a | ± Δ | φ ₁ |.
When the rotation direction is reversed, the same operation is performed after t4.

[0048]

MODIFICATION The present invention is not limited to the above-mentioned embodiments, and the following modifications are possible. (1) The proportional-plus-integral compensation circuit 12 may be a proportional circuit or an integrating circuit. (2) Instead of using only the voltage vector data as the first vector data, the waveform may be improved by using a combination of the voltage vector data and the zero vector data. That is, the zero vector may be arranged in the array of voltage vectors of the memories M1 and M3. (3) Each circuit of FIGS. 1 to 4 can be an individual circuit or an analog circuit without using a DSP.

[Brief description of drawings]

FIG. 1 is a block diagram showing a motor speed control circuit of an embodiment.

FIG. 2 is a block diagram showing a rotational angular velocity calculation and magnetic flux control circuit of FIG.

3 is a block diagram showing in detail the primary magnetic flux calculating circuit and the secondary magnetic flux calculating circuit of FIG.

FIG. 4 is a block diagram showing in detail the ω _m arithmetic circuit of FIG.

5 is a diagram showing in principle a part of the contents of the ROM of FIG.

FIG. 6 is a diagram showing voltage vectors.

FIG. 7 is a diagram showing a rotating magnetic field vector.

FIG. 8 is a diagram showing a state of each part of FIG.

FIG. 9 is a diagram showing a relationship between a magnetic flux change and a vector.

FIG. 10 is a waveform diagram showing the calculated value of expression (1) and the values of its numerator and denominator.

11 is a waveform diagram showing the actual speed, the calculated speed, the load torque, and the secondary resistance of the device of FIG.

[Explanation of symbols]

 1 Motor 2 Inverter 5 ROM 50 In rotational angular velocity calculation / flux control circuit

[Equation 6]

Claims (1)

[Claims] 1. A device for detecting the rotational speed of a three-phase AC motor driven by a three-phase PWM inverter, wherein the three-phase output voltage instantaneous values ( ^v1a , _v1b , v) of the three-phase PWM inverter are detected. voltage detecting means for detecting _1c), the three-phase three-phase output current instantaneous value of the PWM inverter (i _1a, i _1b, a current detecting means for detecting the i _1c), the three-phase output voltage instantaneous value (v _1a , V _1b , v _1c ) are converted into two-phase voltage instantaneous values (v _1d , v _1q ) and the three-phase output current instantaneous values (i _1a , i _1b , i _1c ) are converted into two-phase current instantaneous values (i _1d , I _1q ), and based on the two-phase output voltage instantaneous values (v _1d , v _1q ) and the two-phase output current instantaneous values (i _1d , i _1q ), the primary flux of the AC motor ( φ _1d , φ _1q ) and secondary magnetic flux (φ _2d , φ _2q )
Then, the instantaneous values of the two-phase output current (i _1d , i _1q ), the temporary magnetic flux (φ _1d , φ _1q ), the secondary magnetic flux (φ _2d , φ _2q ), and the primary and secondary AC _motors are obtained. Next and mutual inductance (L ₁₁ ,
L _22, M), characterized in that it consists of a rotational speed calculating means for calculating a value corresponding to the rotational speed of the induction motor based on an AC motor - motor rotational speed detecting device.