JP7532297B2 - Motor Drive Unit - Google Patents

Description

An embodiment of the present invention relates to a drive device that controls a motor without using a rotational position sensor.

In recent years, with the trend toward energy conservation, for example, power conversion devices for driving three-phase brushless DC motors are configured with three half-bridge circuits connected in parallel between the positive and negative DC power lines. The half-bridge circuit consists of a pair of semiconductor switching elements connected in series between the DC power lines, and freewheeling diodes connected in inverse parallel to each of the semiconductor switching elements.

In the power conversion device configured as above, the drive of each semiconductor switching element is controlled by PWM (Pulse Width Modulation). This converts the DC power provided from the DC power supply line into three-phase AC power, and current is passed through the motor windings. As a circuit for driving a brushless DC motor, sensor drive using a position sensor has been proposed, particularly for applications requiring starting torque. However, in order to achieve low cost and compact size, a motor drive device that controls a motor without using a position sensor is required.

For example, instead of detecting the rotational position of the motor with a magnetic pole sensor such as a Hall element, there is a method of detecting the rotational position using the induced voltage generated in the motor windings and outputting the motor voltage based on this. In 120° rectangular wave drive, which outputs a rectangular wave voltage with an electrical angle of 120° for one phase, the induced voltage can be detected in the 60° section where the output stops. However, 120° rectangular wave drive has the problem of generating greater noise and vibration than 180° sine wave drive, which outputs a sine wave voltage.

A common method for sinusoidal wave drive without using a position sensor is to calculate the induced voltage based on the motor voltage equation and use that induced voltage to estimate the rotational position. However, this method imposes a high computational load, requires the use of an expensive microcomputer, and requires motor constants, etc.

On the other hand, there is a technique called V/f control, which keeps the ratio (V/f) of the voltage (V) output from the inverter to the frequency (f) constant. This technique has a low computational load and does not require detection of the rotational position or motor constants.

JP 2009-124812 A

However, because V/f control does not grasp the rotational position of the rotor, if the load torque becomes too heavy, the motor rotation cannot follow the frequency command, resulting in an abnormal stop, i.e., loss of synchronization. On the other hand, if the voltage command is set to a large value to avoid loss of synchronization, efficiency will decrease.
Therefore, the present invention provides a motor drive device that can achieve both high efficiency and robustness against load fluctuations.

The motor drive device of this embodiment includes:
an electric current supply unit that supplies electric current to a stator winding of the motor;
a current detection unit that detects a current flowing through the stator winding;
a peak current detection unit for detecting a peak value of the current;
a phase difference detection unit that detects a phase difference between a voltage applied to the motor and the current;
a voltage calculation unit that calculates the magnitude of the applied voltage based on an input frequency command;
a voltage correction unit that corrects the applied voltage so as to reduce the peak current as much as possible;
a lower limit setting unit that sets a lower limit value of the phase difference;
and a control unit that, when it is determined that the phase difference falls below the lower limit, reduces the frequency command and increases the correction voltage to be applied to the voltage correction unit.

FIG. 1 is a functional block diagram showing a configuration of a motor drive device according to an embodiment. Vector diagram for V/f control A graph showing Figure 2 (part 1) A graph showing Figure 2 (part 2) FIG. 13 is a diagram showing the relationship between the voltage/current phase difference and torque when the motor rotation speed is 2900 rpm, 3000 rpm, and 3100 rpm with a duty of 90%. A diagram showing the control waveform when a load is applied without "speed command adjustment" A diagram showing the control waveform when a load is applied with "speed command adjustment" A diagram showing how the lower limit of the phase difference is changed according to a frequency command. FIG. 13 is a diagram showing the effect of the output voltage correction amount calculation unit. A diagram comparing the efficiency of the V/f control and the vector control of this embodiment. A diagram showing the state in which output voltage correction is being performed stably. FIG. 12 shows a state in which the frequency command is halved from the state shown in FIG. 11 . FIG. 13 is a diagram showing a state in which the correction amount of the output voltage is made proportional to the frequency command in the case shown in FIG. 12 . A diagram showing the state in which the frequency command is adjusted to reduce the output voltage, and then the frequency command is repeatedly accelerated and decelerated. FIG. 15 is a diagram showing waveforms when “output voltage adjustment” is introduced into the case shown in FIG. 14 . FIG. 13 is a diagram showing a determination of out-of-step according to a change in the voltage-current phase difference. Flowchart of the main control routine Flowchart of output adjustment process

An embodiment will be described below with reference to the drawings. FIG. 1 is a functional block diagram showing the configuration of a motor drive device. Each phase output terminal of an inverter 2, which is a power converter, is connected to each phase winding of a motor 1, which is, for example, a permanent magnet motor. Although not specifically shown, the inverter 2 is configured with a half-bridge circuit connected in parallel for three phases between positive and negative DC power supply lines. The half-bridge circuit is composed of a pair of semiconductor switching elements connected in series between the DC power supply lines, and a free wheel diode connected in inverse parallel to each of the semiconductor switching elements. The semiconductor switching elements are, for example, N-channel MOSFETs.

The frequency command acceleration/deceleration limiting unit 3 receives a frequency command ωref from the outside, and applies acceleration/deceleration limits to the frequency command ωref as necessary, and outputs the limit to the f/V conversion unit 5 via the subtractor 4. For example, if the frequency command ωref suddenly changes in a step shape, the acceleration/deceleration limiting unit 3 converts the command value so that it increases gradually in proportion to time. The f/V conversion unit 5 outputs a voltage signal converted according to the input frequency command as a duty command in PWM control. The frequency command is also converted into a position command by being integrated by the integrator 6.

The duty command and position command are input via the subtractor 8 to the modulation control unit 7, which generates a three-phase PWM signal (U, V, W±) and outputs it as a gate signal to each semiconductor switching element of the inverter 2.

The voltage-current phase difference detection processing unit 9 detects the motor current from the voltage from a single shunt resistor placed only one on the negative power supply side of the inverter 2, or a triple shunt resistor placed on the negative power supply side of each phase arm. The voltage-current phase difference detection processing unit 9 calculates the difference between the voltage phase obtained from the modulation control unit 7 and the motor current phase. In the case of the triple shunt method, the phase difference is calculated at the zero crossing point of the current, and in the case of the single shunt method, the phase difference is calculated by finding the zero crossing point of the current of the remaining phase from the crossing timing of the two-phase current.

The current peak detection processing unit 10 finds the peak from the detected motor current, and the output voltage correction amount calculation unit 11 outputs a correction value to the subtractor 8 for correcting the output voltage so that the current peak is minimized. When the phase difference detected by the voltage-current phase difference detection processing unit 9 falls below the lower limit input by the lower limit setting unit 14, the frequency command correction amount calculation unit 12, which corresponds to the control unit, outputs a correction value to the subtractor 4 that reduces the frequency command, and outputs a command value to the output voltage correction amount calculation unit 11 that increases the duty command.

The out-of-step determination unit 13, which corresponds to the rotation abnormality determination unit, determines that the motor 1 has gone out of step based on the change in the voltage-current phase difference detected by the voltage-current phase difference detection processing unit 9. The correction timing determination unit 15 causes the output voltage correction amount calculation unit 11 to start voltage correction processing when the frequency command ωref and the output of the frequency command acceleration/deceleration limiting unit 3 match. The above constitutes the motor control device 16.

Figure 2 shows a vector diagram during V/f control. When the load torque is small, the current vector I is close to the d-axis, but as the load increases, it moves closer to the q-axis to increase the torque. If the load becomes too large, field-weakening control is performed, and the current increases, increasing the torque, but if the voltage phase advances too much, the torque decreases, resulting in loss of synchronization.

Figures 3 and 4 are graphs of Figure 2, showing the relationship between the phase difference between the voltage and current and the torque and current. When the PWM duty is constant, as the load increases, the magnetic pole lags, the d-axis current Id becomes negative, the voltage phase advances, and the phase difference becomes smaller.

Figure 5 shows the relationship between the voltage-current phase difference and torque when the duty is fixed at 90% and the motor rotation speed is 2900 rpm, 3000 rpm, and 3100 rpm. Based on this relationship, in this embodiment, when the phase difference becomes small and falls below 0 [deg], a control is added to reduce the speed command, i.e., the frequency speed command, to increase the phase difference.

Figure 6 shows the control waveforms when a load is applied "without speed command adjustment," and Figure 7 shows the control waveforms when "with speed command adjustment." Without speed command adjustment, the motor stepped out of step and stopped, but with speed command adjustment, the speed command is lowered in the area circled in the figure, so stopping due to the motor stepping out can be avoided. In addition, by changing the lower limit value of the phase difference so that it is increased in response to the frequency command ωref, for example as shown in Figure 8, the lower limit setting unit 14 can ensure robustness against load fluctuations even if the frequency command ωref changes.

Figure 9 shows the effect of the output voltage correction amount calculation unit 11. The voltage command, i.e., the duty command, is automatically adjusted so that the peak value of the motor current is reduced. Details of the processing of the output voltage correction amount calculation unit 11 will be explained in the control flowchart described later.

Figure 10 compares the efficiency of the V/f control of this embodiment with that of vector control. The efficiency difference with vector control is less than one percentage point, and the control of this embodiment achieves performance equivalent to that of vector control.

The amount of output voltage correction must be changed according to the frequency command. In Figures 11 to 13, in "voltage correction execution," correction is performed at the timing when the waveform rises in an impulse shape. For example, if the frequency command is halved as shown in Figure 12 from a state in which output voltage correction is being performed stably as shown in Figure 11, the amount of correction becomes too large, making control unstable and making it difficult to suppress the current peak value. In contrast, by making the amount of output voltage correction proportional to the frequency command as shown in Figure 13, output voltage correction can be performed well even if the frequency command suddenly drops, and control is stable and the current peak value can be appropriately suppressed.

Although the output voltage is reduced by performing the correction as described above, as shown in FIG. 14, after the frequency command is adjusted and reduced, there is a problem that the output voltage for accelerating the motor 1 is insufficient and the original frequency command cannot be restored, resulting in repeated acceleration and deceleration of the frequency command. Therefore, in this embodiment, by increasing the output voltage when the frequency command is reduced, both "output voltage adjustment" that realizes high efficiency in V/f control and "frequency command adjustment" that responds to load fluctuations are achieved. FIG. 15 shows the waveform when "output voltage adjustment" is introduced, and the state in which repeated acceleration and deceleration of the frequency command are avoided.

Here, if the correction voltage calculated by the output voltage correction amount calculation unit 11 is used as is when the frequency command is lowered, the correction amount will be too large, causing the output voltage to become a negative value, and the motor 1 will not be able to drive. To solve this problem, the correction voltage calculated by the voltage correction unit can be reduced in accordance with the proportion of the lowered frequency command relative to the original frequency command before the reduction. This prevents overcorrection of the output voltage and prevents the motor 1 from stopping.

In addition, because V/f control does not detect the rotor rotation position of motor 1, it is necessary to properly detect when a sudden change in the load causes a loss of synchronism. In this embodiment, as shown in the area surrounded by an ellipse in Figure 16, loss of synchronism is detected by determining that a loss of synchronism has occurred when the voltage-current phase difference suddenly changes significantly to a positive value.

Figure 17 shows a flowchart of the main control routine. This process is executed for each PWM period, for example, with a frequency of about 20 kHz. First, the correction timing determination unit 15 determines whether the rotation frequency ω of the motor 1 matches the frequency command ωref (S01). If the rotation frequency ω does not match the frequency command ωref (NO), the rotation frequency is increased or decreased (S02). The increase or decrease here is performed according to the frequency command increase/decrease unit 3.

When the rotation frequency ω matches the frequency command ωref (S01; YES), output voltage correction is started, and the voltage-current phase difference detection processing unit 9 and the peak detection processing unit 10 capture the current value. If the current current value is greater than the held peak value, the peak detection processing unit 10 holds the updated current peak value (S1).

When the voltage/current phase difference detection processing unit 9 detects the zero crossing point of the current (S2; YES), it executes processing for one motor electrical angle cycle. First, it saves the previously calculated voltage/current phase difference (S3) and then calculates the current phase difference (S4). If the amount of change in the phase difference between the previous and current phases is greater than threshold 1 (S5; YES), the out-of-step determination unit 13 determines that out-of-step has occurred (S8). If the current phase difference is equal to or less than threshold 2 (S6; YES), the load on the motor 1 is large, so the frequency command is reduced (S9) and the output voltage is increased (S10). If the current phase difference is greater than threshold 2 (S6; NO), output adjustment processing is performed (S7).

In the output adjustment process shown in FIG. 18, an index ε is calculated from the current peak value Ip or the current peak value Ip and the frequency ω, and an integrated value γa of the current peak value or the index ε is calculated (S11). The frequency ω uses the frequency command ωref. Here, the integrated value γa is reset to zero every αref cycles of the current I by the process described below. αref is a natural number equal to or greater than 2. Therefore, the integrated value γa is also reset to zero every several cycles of the current I.

The integrated value γa(n) in the nth period is expressed by equation (1).
γa(n)=γa(n-1)+ε(n)=γa(n-1)+Ip/ω...(1)
Ip: current peak value acquired in the nth period
ω: Frequency obtained in the nth period
ε(n): Index of the nth period γa(n-1): Integrated value of the (n-1)th period

Next, the output voltage correction amount calculation unit 11 judges which stage (0 to 4) the output adjustment situation of the drive voltage V is in (S12). In the first judgment, for example, it is judged that the stage is 0.
<When Stage=0: Decrease the output voltage>
The output voltage correction amount calculation unit 11 reduces the drive voltage V by a predetermined voltage amount (S13), sets Stage=1 (S14), and increments the count value α (S15). The count value α indicates how many periods of the current I have passed since the previous comparison using the integrated value γa.

<When Stage=1: The output voltage is maintained as is>
The output voltage correction amount calculation unit 11 judges whether the count value α is equal to or greater than a reference value αref (S16). The reference value αref is a periodic value for resetting the integrated value γa to zero. If the count value α is smaller than the reference value αref (NO), the count value α is incremented (S22) and Stage=1 is maintained.

On the other hand, if the count value α is equal to or greater than the reference value αref (S16; YES), the output voltage correction amount calculation unit 11 compares the integrated value γa from the previous comparison to the current comparison with the integrated value γb from the comparison before last to the previous comparison (S17). If the integrated value γa is smaller than the integrated value γb (S18; YES), Stage is set to 0 (S19). On the other hand, if the integrated value γa is equal to or greater than the integrated value γb (NO), Stage is set to 2 (S21). Then, the count value α and the integrated value γa are reset to zero, and the integrated value γb is updated with the integrated value γa (S20).

<When Stage=2: Increase the output voltage>
The output voltage correction amount calculation unit 11 increases the drive voltage V by a predetermined voltage amount (S23), sets Stage=3 (S24), and increments the count value α (S25).

<Stage=3: Maintain the output voltage as is>
The output voltage correction amount calculation unit 11 judges whether the count value α is equal to or greater than the reference value αref (S26). If the count value α is smaller than the reference value αref (NO), the count value α is incremented (S32) and Stage=3 is maintained. On the other hand, if the count value α is equal to or greater than the reference value αref (YES), the integrated value γa is compared with the integrated value γb (S27). If the integrated value γa is smaller than the integrated value γb (YES), Stage=2 is set (S29). On the other hand, if the integrated value γa is equal to or greater than the integrated value γb (NO), Stage=0 is set (S31). Then, the same process as in step S20 is performed (S30).

As described above, according to this embodiment, in the motor drive device 16, the current peak detection processing unit 10 detects the peak value of the current passed through the motor 1, and the voltage-current phase difference detection processing unit 9 detects the phase difference between the output voltage to the motor 1 and the current. The f/V conversion unit 5 calculates the output voltage Duty based on the input frequency command ωref, and the output voltage correction amount calculation unit 11 corrects the output voltage so as to reduce the peak value of the current as much as possible. Then, when the frequency command correction amount calculation unit 12 determines that the phase difference has fallen below the lower limit, it performs a correction to reduce the frequency command ωref and causes the output voltage correction amount calculation unit 11 to perform the correction.

By configuring in this way, it is possible to prevent stepping out of motor 1 by correcting the frequency command ωref, and to avoid a decrease in the output torque of motor 1 by correcting the output voltage, thereby enabling V/f control to be performed with high efficiency.

The correction timing determination unit 15 also activates the output voltage correction amount calculation unit 11 when the current supply frequency to the motor 1 reaches the frequency command ωref, so that the output voltage correction can be started at an appropriate timing. In this case, the output voltage correction amount calculation unit 11 changes the value by which the correction voltage is increased according to the current supply frequency to the motor 1, so that even if the frequency command suddenly drops, the output voltage is well corrected, and the control is stable and the current peak value can be appropriately suppressed.

Furthermore, the output voltage correction amount calculation unit 11 adjusts the value of the correction voltage based on the frequency command ωref, thereby preventing overcorrection of the output voltage and preventing the motor 1 from stopping. In addition, the lower limit setting unit 14 changes the lower limit of the phase difference according to the frequency command ωref, ensuring robustness against load fluctuations even if the frequency command ωref changes.

In addition, when the out-of-step determination unit 13 determines that the amount of change between the current phase difference and the previous phase difference exceeds a threshold value, it determines that there is an abnormality in the rotation of the motor 1, i.e., out-of-step. Therefore, even in V/f control that does not detect the rotor position, it is possible to reliably determine out-of-step of the motor 1.

Other embodiments
The control device 16 may be configured as a motor driver IC (integrated circuit).
The lower limit value of the phase difference does not necessarily need to be changed in response to the frequency command ωref, and may be a constant value.

Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are included in the scope of the invention and its equivalents described in the claims.

In the drawing, 1 is a motor, 2 is an inverter, 5 is an f/V conversion unit, 9 is a voltage-current phase difference detection processing unit, 10 is a current peak detection processing unit, 11 is an output voltage correction amount calculation unit, 12 is a frequency command correction amount calculation unit, 13 is an out-of-step determination unit, 14 is a lower limit value setting unit, 15 is a correction timing determination unit, and 16 is a motor control device.

Claims (6)

an electric current supply unit that supplies electric current to a stator winding of the motor;
a current detection unit that detects a current flowing through the stator winding;
a peak current detection unit for detecting a peak value of the current;
a phase difference detection unit that detects a phase difference between a voltage applied to the motor and the current;
a voltage calculation unit that calculates the magnitude of the applied voltage based on an input frequency command;
a voltage correction unit that corrects the applied voltage so as to reduce the peak current as much as possible;
a lower limit setting unit that sets a lower limit value of the phase difference;
a control unit that, when it is determined that the phase difference falls below the lower limit, reduces the frequency command and increases the correction voltage to be applied to the voltage correction unit. The motor drive device according to claim 1, further comprising a correction timing determination unit that activates the voltage correction unit when the current supply frequency to the motor reaches the frequency command. The motor drive device according to claim 1 or 2, wherein the control unit changes the value by which the voltage correction unit increases the correction voltage depending on the frequency of current flow to the motor. The motor drive device according to any one of claims 1 to 3, wherein the control unit adjusts the value of the correction voltage calculated by the voltage correction unit based on the frequency command. The motor drive device according to any one of claims 1 to 4, wherein the lower limit setting unit changes the lower limit in response to the frequency command. The motor drive device according to any one of claims 1 to 5, further comprising a rotation abnormality determination unit that determines that the motor has a rotation abnormality when it is determined that the amount of change between the current phase difference and the previous phase difference exceeds a threshold value.

Images