JP2006014426A - Controller for ac motor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress the offset of a current by reducing the slippage of voltage SW timing caused by the detection error of a position detector, in a controller for an AC motor which performs rectangular wave voltage drive. <P>SOLUTION: This controller for an AC motor computes the electric angular velocity ω, using the electric angle θ detected with a position detector 6, and divides the reference phase difference from the start of one voltage SW pattern to the end by the electric angular velocity ω thereby converting it into a reference phase difference time t'. The controller divides the phase difference (θsw<SP>*</SP>-θ next) between the electric angle target value θsw<SP>*</SP>in the switching of the voltage SW pattern at the next control operation and the electric angle predicted value θnext at the next control operation by the electric angle velocity ω thereby converting it into a phase difference error time Δt. The controller also corrects the above reference phase error time with the value being obtained by multiplying this phase error time by a coefficient K in a range larger than 0 and smaller than 1, and sets carrier frequency, according to the value after correction. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a control device for an AC motor, and more particularly to a control device that controls an inverter with a rectangular wave voltage drive.

As a conventional control device for an AC motor, there is one described in Patent Document 1 below.
As a method for controlling the inverter that supplies current to the motor by such an AC motor control method, the PWM voltage drive method includes a rectangular wave voltage drive method that is used in an operation region where the output voltage is limited. In the rectangular wave voltage driving method described in Patent Document 1, a timer counter that operates with a clock at a fixed interval is used, and a voltage switching pattern (hereinafter abbreviated as a voltage SW pattern) is switched every time the count value reaches a target value. Is output. The phase difference target value from the start to the end of the voltage SW pattern is obtained as an ideal reference phase difference from the start to the end of one section of the voltage SW pattern, and the reference phase difference is determined by the phase error based on the deviation between the target torque and the estimated torque. It is obtained by correcting. The counter target value is determined according to a value obtained by time-converting the phase difference target value, and a position detector for obtaining the electrical angle θ is unnecessary.

JP 2002-359996 A

As described above, in the conventional example, the phase difference target value is calculated by correcting the phase difference reference value with the phase error based on the torque deviation. When performing control such that the phase difference target value is determined based on the detection amount of the detector, the voltage SW timing may be shifted due to the detection error of the position detector, and the current may be offset. was there.
The present invention has been made to solve the above-described problem, and in the rectangular wave voltage drive control, an AC motor in which a deviation in voltage SW timing caused by a detection error of a position detector is reduced and a current offset is suppressed. An object of the present invention is to provide a control device.

In order to achieve the above object, in the present invention, the electrical angular velocity ω is calculated using the electrical angle θ detected by the position detector, and the reference phase difference from the start to the end of one voltage SW pattern is calculated as the electrical angular velocity ω. Is converted into a reference phase difference time t ′, and the phase error (θsw) between the electrical angle target value θsw ^{* for} switching the voltage SW pattern at the next control calculation and the predicted electrical angle θnext at the next control calculation is calculated. ^* −θnext) is divided by electrical angular velocity ω to convert to phase error time Δt, and the reference phase difference time is corrected by a value obtained by multiplying this phase error time by a coefficient K in a range greater than 0 and less than or equal to 1. The carrier period is set in accordance with the corrected value.

  By appropriately setting the value of the coefficient K, it is possible to reduce the amount that the error of the position detector appears as the error of the voltage SW pattern switching timing, thereby reducing the deviation of the switching timing of the voltage SW pattern due to the error of the position detector, There is an effect that current offset can be suppressed.

FIG. 1 is a block diagram of an embodiment showing a configuration of an AC motor control apparatus to which the present invention is applied.
In FIG. 1, the voltage phase generating means 1 outputs a voltage phase target value α ^* obtained by referring to a table using an externally input torque command value T ^* and the current rotational speed ω as indices. Specifically, for example, in the evaluation test of the electric motor to be controlled, the value of the voltage phase target value α ^* with respect to the torque command value T ^* and the rotational speed ω is obtained as table data. The value of the voltage phase target value α ^* corresponding to the torque command value T ^* and the rotation speed ω can be obtained by referring to the table.

  The control unit 2 includes a rectangular wave control unit 2-1 and a PWM control unit 2-2. Since the present invention relates to rectangular wave voltage driving, the rectangular wave control means 2-1 will be mainly described below, and only necessary portions of the PWM control means 2-2 will be described.

The rectangular wave control means 2-1 determines the voltage phase target value α ^* output from the voltage phase generation means 1, the electrical angle θ of the electric motor 5 detected by the position detector 6 (for example, a resolver), and the electrical angle θ. The electric angular velocity ω (rotational speed) obtained by the velocity calculating means 7 as an input is input, and the drive signal P of the on / off signal is calculated and output (details will be described later). The inverter 3 is controlled by this drive signal P, and three-phase rectangular wave voltages Vu, Vv, and Vw having an amplitude of the power supply voltage Vdc or 0 (or + Vdc / 2 or −Vdc / 2) are output from the inverter 3. To drive the three-phase motor 5.

The voltage phase generation means 1 and the control means 2 are constituted by a computer or the like, and calculate the drive signal P by repeatedly performing calculations at a predetermined cycle.
The drive signal P is an on / off signal, and the output of the inverter 3 is output in synchronization with the drive signal P. That is, the timing at which the drive signal P is switched on / off is the timing at which the rectangular wave voltage is switched as it is (strictly speaking, on and off may be reversed). In the rectangular wave voltage drive, the applied voltage amplitude is the power supply voltage Vdc or 0 (+ Vdc / 2 or −Vdc / 2) and the amplitude cannot be controlled, so that the voltage phase follows the voltage phase target value α ^*. By controlling, the torque control of the electric motor 5 is performed so as to realize the given torque command value T ^* . That is, since there is a correlation between the torque and the voltage phase, the torque can be controlled by controlling the voltage phase. This voltage phase is controlled by switching a voltage SW pattern described later.

  Note that the PWM control means 2-2 in the control means 2 performs a general torque control calculation using the detected currents Iu and Iv detected by the current sensor 4 in the region where the PWM voltage drive is performed, and the inverter 3 outputs the PWM signal. The torque is controlled by changing the voltage value applied to each phase of the electric motor 5. The general torque control calculation in the above-described PWM control is, for example, calculating a d-axis current command value and a q-axis current command value based on the input torque command value and the rotation angle of the electric motor 5, and d-axis current command The proportional-integral calculation is performed based on the deviation between the value and the actual d-axis current value to calculate the d-axis voltage command value, and the q-axis is similarly calculated based on the deviation between the q-axis current command value and the actual q-axis current value. Calculate the voltage command value. The actual d-axis current value and q-axis current value are obtained by performing three-phase to two-phase conversion from the detected currents Iu and Iv (Iw can be calculated from Iu and Iv). Then, the d-axis voltage command value and the q-axis voltage command value are two-phase / three-phase converted to calculate a three-phase voltage command value. The drive signal P is obtained by calculating the duty command value of the PWM signal from the three-phase voltage command value and comparing the duty command value with a predetermined carrier signal (such as a triangular wave or a sawtooth wave).

  When the rectangular wave voltage drive used in the present invention is performed, the rectangular wave (Vdc or 0) is set by setting the duty ratio of the duty command value to be compared with the carrier signal to 0 [%] or 100 [%]. Drive signal P can be generated. In general, the rectangular wave voltage drive is used in a field weakening region where a high voltage is required, and PWM control is used in other regions.

In addition, in order to control the voltage phase to follow the voltage phase target value α ^* in the rectangular wave control means 2-1, the voltage SW pattern (detailed later) is determined at a timing determined according to the voltage phase target value α ^*. Do by switching.
The setting of the voltage SW pattern and the carrier cycle is effective at the trough of the triangular wave or sawtooth wave of the carrier signal, and at the same time, an interrupt for starting the control calculation is generated. The timing at which the rectangular wave pattern (voltage SW pattern) is switched is adjusted by changing the carrier cycle so that the voltage SW pattern switching timing is appropriate. In other words, the voltage phase is set to voltage by controlling the period of the carrier signal (for example, triangular wave) for creating the rectangular wave so that the switching point of the voltage SW pattern coincides with the valley of the carrier signal (at the start of the interrupt calculation). Control is performed to follow the phase target value α ^* .

(Example)
Hereinafter, the carrier period setting method in the embodiment will be described in detail.
FIG. 2 is a signal waveform diagram showing a carrier period setting method.
In FIG. 2, the control calculation is interrupted and the voltage SW pattern is switched at the valleys (time points t ₀ and t ₁ ) of the carrier signal (triangular wave). The voltage SW pattern is a voltage pattern applied to each of the three phases U, V, and W, that is, a pattern of which one of the three phases is set to “1”, that is, Vdc, and which is set to “0”. ₀ in ~t ₁ is in a "Vu = 1 (Vdc), Vv = 0, Vw = 0 ".

In the control calculation 1 performed at the time point t ₀ , the carrier cycle Tnext of one section of the next voltage SW pattern is used as follows using the electrical angle θ and the electrical angular velocity ω at the start of the control calculation 1 and the current carrier cycle Tow. To calculate.

First, as shown in (Equation 1), a time t ′ (reference phase difference time) is obtained by dividing the reference phase difference π / 3 [rad] from the start to the end of one voltage SW pattern by the electrical angular velocity ω. Convert to.
t ′ = π / 3ω (Equation 1)
Next, as shown in the equation (2), an electrical angle predicted value θnext at the time of switching the next voltage SW pattern is calculated.
θnext = θ + ωTnow (Equation 2)
The current carrier cycle Tow corresponds to the next predicted electrical angle value θnext calculated in the previous control calculation.

The electrical angle target value of the next voltage SW pattern switching .theta.sw ^*, based on FIG. 3 showing the electrical angle theta, the voltage phase target value alpha ^*, the relationship between the SW pattern of voltage to be input to the motor, θsw0 ^* ~ The electrical angle closest to θnext is selected from θsw5 ^* , and the voltage SW pattern output after the next voltage SW pattern switching is determined from the current rotation direction. For example, in FIG. 3, if θnext is the closest value to (2π / 3) −α ^* , θsw2 ^* is selected as the electrical angle target value θsw ^* for the next voltage SW pattern switching, and the next voltage SW The pattern is “Vu = −Vdc / 2, Vv = −Vdc / 2, Vw = + Vdc / 2”.

  In FIG. 3, each voltage SW pattern is a binary rectangular wave of + Vdc / 2 and −Vdc / 2 with 0 as the center. This is because, when the three-phase winding of the motor is Y-connected, any two of the three phases U, V, and W are connected in series between the power supply voltage Vdc terminal and the ground terminal. If the neutral point is set to 0, + Vdc / 2 is applied to the phase connected to the Vdc terminal side, and -Vdc / 2 is applied to the phase connected to the ground terminal side. It is because it can be displayed. Therefore, as shown in FIG. 2, + Vdc / 2 may be represented by “1” (Vdc) and −Vdc / 2 may be represented by “0”.

As described above, the electrical angle target value of the next voltage SW pattern switching θsw ^*, θsw0 ^* ~θsw5 ^* calculated by selecting those electrical angle prediction value θnext is closest to the phase error a difference θnext and .theta.sw ^* As shown in the following (Equation 3), the phase error (θsw ^* −θnext) is divided by the electrical angular velocity ω to be converted into time Δt (phase error time).
Δt = (θsw ^* −θnext) / ω (Equation 3)
Next, as shown in the following (Expression 4) and (Expression 5), t ′ obtained by converting the reference phase difference π / 3 into time is set to the above Δt with a coefficient K (1 greater than 0). A time t is corrected by a value multiplied by a value in the following range), and this is set as the next carrier cycle Tnext.
t = t ′ + K · Δt (Equation 4)
Tnext = t (Equation 5)
Further, the next voltage SW pattern is selected based on FIG. 3 from the electrical angle target value θsw ^* of the next voltage SW pattern switching and the rotation direction.

If the phase error (θsw ^* −θnext) is corrected by Δt converted to time as described above, the next voltage SW pattern switching timing can be controlled to coincide with the valley of the carrier signal. Therefore, the coefficient K in the above equation (4) may be set to K = 1 in principle. However, as described above, the phase error time Δt is based on the electrical angle θ detected by a position detector such as a resolver, and the detection value of the position detector includes an error as described above. Therefore, Δt is also a value including an error of the position detector. If correction for Δt is performed at a time with K = 1, the voltage SW pattern switching timing may be shifted. Therefore, the value of the coefficient K is set to a value as small as possible (for example, 0.1) within the range of greater than 0 and less than 1 (actually less than 1), and the phase error is sequentially corrected by a plurality of operations. However, the control performance is improved.

Hereinafter, the above problem will be described.
4A and 4B are diagrams showing an example of the position detector error and the deviation of the voltage SW pattern caused by the error. FIG. 4A is a diagram showing the error characteristic of the position detector, and FIG. 4B is a reference target voltage waveform. (C) is an actual voltage waveform diagram in the case of K = 1, and (d) is an actual voltage waveform diagram in the case of K = 0.1.

As shown in FIG. 4A, in the position detector (for example, resolver), the detected electrical angle indicated by the solid line 2 is substantially the same as the electrical angle period with respect to the actual electrical angle θ indicated by the broken line 1. It has a sinusoidal error.
When the electrical angle θ detected by the position detector includes such an error, as shown in the target voltage waveform 3 of FIG. 4B, the voltage is obtained when the electrical angle is 0 [rad] and π [rad]. When it is desired to switch the SW pattern, if the coefficient is set to K = 1 and correction for Δt is performed at a time, the detection errors Δθ ₁ and Δθ ₂ of the position detector are directly used as the switching timing. Since it appears as an error, the voltage SW pattern switching electrical angle is shifted by detection errors Δθ ₁ and Δθ ₂ as shown in the actual voltage waveform 4 of FIG.

On the other hand, when K = 0.1, the voltage SW pattern switching electrical angle is shifted by 1/10 of the detection errors Δθ ₁ and Δθ ₂ , so that the actual voltage waveform 5 in FIG. As shown in FIG. 5, the deviation of the voltage SW pattern switching electrical angle due to the detection errors Δθ ₁ and Δθ ₂ becomes small.

  As shown in FIGS. 4C and 4D, the current offset increases as the unbalance between the lengths of the ON and OFF periods of the voltage in the voltage SW pattern increases. Therefore, by reducing K as much as possible, It is possible to suppress an unbalance between the lengths of the voltage ON section and the OFF section and reduce the current offset. However, if the coefficient K is decreased, the number of operations until the deviation of the voltage SW pattern switching electrical angle is corrected increases. For example, when K = 0.1, the deviation that can be corrected by one control calculation in a steady state is 1/10. Therefore, ten control calculations are required to correct all the deviations. Therefore, an optimum value (for example, about 0.1) is set in the range of 0 <K ≦ 1 in consideration of the balance with the detection error of the position detector. When the value of K is reduced, it is difficult to perform sufficient correction in a transient state such as during sudden acceleration / deceleration, but the voltage SW pattern switching electrical angle shifts during sudden acceleration / deceleration. Since it is unavoidable to occur to some extent, it is determined in consideration of the stability of control in a steady state.

The block diagram of one Example which shows the structure of the control apparatus of the alternating current motor to which this invention is applied. FIG. 4 is a signal waveform diagram for explaining a carrier period setting method in the first embodiment. The figure which shows SW pattern of the voltage which should be input into an electric motor. The figure which showed an example of the error of a position detector, and the shift | offset | difference of the voltage SW pattern by it.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Voltage phase generation means 2 ... Control means 2-1 ... Rectangular wave control means 2-2 ... PWM control means 3 ... Inverter 4 ... Current sensor 5 ... Electric motor 6 ... Position detector 7 ... Speed calculation means

Claims (1)

In a control device for an AC motor that performs rectangular wave voltage drive that applies the highest value and the lowest value of the power supply voltage to each phase of the motor winding in a predetermined voltage switching pattern,
Voltage phase generation means for outputting a voltage phase target value according to a torque command value given from the outside and the current rotation speed;
A position detector for detecting the electrical angle of the electric motor;
Rectangular wave control means for calculating a drive signal of a rectangular wave for driving the electric motor from the voltage phase target value, the electric angle of the electric motor, and the electric angular velocity of the electric motor;
An inverter that drives the electric motor by applying a rectangular wave voltage corresponding to the rectangular wave driving signal to each phase of the motor winding; and
The rectangular wave control means repeatedly generates an interrupt for starting a control calculation for each valley position of the carrier signal and performs setting of the voltage switching pattern and setting of a carrier period of the carrier signal,
In addition, the electrical angular velocity is calculated from the electrical angle detected by the position detector, and the reference phase difference from the start to the end of one voltage switching pattern is divided by the electrical angular velocity to be converted into the reference phase difference time. By dividing the phase error between the electrical angle target value for switching the voltage switching pattern during the control calculation and the predicted electrical angle value during the next control calculation by the electrical angular velocity, the phase error time is converted to zero. A control apparatus for an AC motor, wherein the reference phase difference time is corrected by a value obtained by multiplying a coefficient in a range of 1 or less and a next carrier cycle is set according to the corrected value. .

Images