JP2024069063A - Magnetic pole position estimation device and magnetic pole position estimation method - Google Patents

Abstract

[Problem] To provide a magnetic pole position estimation device that can reduce shock and vibration to a mechanical system and estimate a rotor magnetic pole position with high accuracy in a small number of steps, even when a synchronous motor is used in a mechanical system where static friction or Coulomb friction exists. [Solution] The magnetic pole position estimation device includes a current detector 9 that detects a current flowing through a synchronous motor, a feedback side coordinate converter 10 that performs coordinate conversion of the current detection value detected by the current detector 9, a current control unit that calculates a voltage command from a current command and a current feedback coordinate-converted by the feedback side coordinate converter 10, a command side coordinate converter 6 that performs coordinate conversion of the voltage command or the current command, a position detection unit 11 that detects the rotor position, a magnetic pole position calculation unit 12 that calculates a control magnetic pole position or a magnetic pole deviation amount based on at least one of the rotor position, the position command, the rotor speed, and the speed command, and an acceleration calculator 15 that calculates the acceleration of the rotor. [Selected Figure] FIG.

Description

The present invention relates to a magnetic pole position estimation device and a magnetic pole position estimation method.

In a synchronous motor that generates torque by passing a current through the stator windings according to the rotor's magnetic pole position, a technique is known that uses an incremental encoder as a rotor position detector to estimate the initial value of the rotor's magnetic pole position when driving begins.

For example, Non-Patent Document 1 discloses a technique for estimating an initial value of the magnetic pole position of the rotor when driving is started. Specifically, if there is a magnetic pole position error, which is a deviation between the magnetic pole position of the rotor and the control magnetic pole position when a constant dq axis current flows, a torque is generated according to the magnetic pole position error, and the rotor position moves. The moved rotor position is detected, and the deviation from the rotor position before the movement is calculated by PI to correct the coordinate transformation position, thereby converging the magnetic pole deviation amount to 0 and estimating the initial value of the magnetic pole position of the rotor.
Patent Literature 1 also discloses a technique for estimating an initial value of a magnetic pole position of a rotor without being affected by static friction of a mechanical system using a synchronous motor. Specifically, in addition to the technique disclosed in Non-Patent Literature 1, an average value of convergence values of magnetic pole deviation amounts when the rotor position is moved in a positive direction and when it is moved in a negative direction is calculated to estimate an initial value of the magnetic pole position of the rotor.

Institute of Electrical Engineers of Japan, Journal of Electrical Engineering, Vol. 122, No. 9, 2002 "Method of detecting magnetic pole position and control of brushless DC servo motor with incremental encoder"

JP 2009-183022 A

However, the amount of current flowing through the stator winding is limited by the inverter capacity of the motor control device. Therefore, when the static friction or Coulomb friction of the mechanical system using a synchronous motor is large, the method of Non-Patent Document 1 cannot reduce the magnetic pole position error, making it difficult to estimate the magnetic pole position of the rotor with high accuracy. In addition, the method of estimating the initial value of the magnetic pole position of the rotor disclosed in Patent Document 1 is composed of four steps, and there is a problem that the estimation time is long. Furthermore, the mechanical system is composed of a motor, a power transmission mechanism, a table, etc., and for example, a workpiece or a transported object is placed on the table, and it is often necessary to prevent impact. The method of Non-Patent Document 1 and the estimation method disclosed in Patent Document 1 apply the q-axis current and the d-axis current in a stepped manner, so that excessive torque is generated when the current is applied, causing an impact on the mechanical system.

The present invention provides a magnetic pole position estimation device that reduces the shock and vibration to a mechanical system and can estimate the rotor's magnetic pole position with high accuracy and a small number of steps, even when a synchronous motor is used in a mechanical system where static friction or Coulomb friction exists.

A magnetic pole position estimation device for estimating a magnetic pole position of a rotor of a synchronous motor according to one aspect of the present invention includes:
a current detector for detecting a current flowing through the synchronous motor;
a feedback side coordinate converter that converts a current detection value detected by the current detector into a coordinate converter;
a current control unit that calculates a voltage command from a current command and the current feedback that has been coordinate-converted by the feedback-side coordinate converter;
a command-side coordinate converter that converts the voltage command or the current command into a coordinate;
a position detection unit for detecting a position of the rotor;
a magnetic pole position calculation unit that calculates a control magnetic pole position or a magnetic pole deviation amount based on at least one of the rotor position and a position command and the rotor speed and a speed command;
an acceleration calculator for calculating an acceleration of the rotor;
Equipped with
the feedback-side coordinate converter and the command-side coordinate converter perform coordinate conversion based on the control magnetic pole position or the magnetic pole deviation amount calculated by the magnetic pole position calculation unit, so that the magnetic pole position estimation device estimates the magnetic pole position of the rotor;
The current command includes a dither signal.
The magnitude of the dither signal is controlled based on the acceleration of the rotor calculated by the acceleration calculator or the speed of the rotor.

A magnetic pole position estimation method for estimating a magnetic pole position of a rotor of a synchronous motor according to one aspect of the present invention includes the steps of:
a current detection step of detecting a current flowing through the synchronous motor;
a feedback side coordinate transformation step of transforming a current detection value detected in the current detection step into a coordinate transformation state;
a current control step of calculating a voltage command from the current command and the current feedback that has been coordinate-converted in the feedback-side coordinate conversion step;
a command-side coordinate conversion step of converting the voltage command or the current command into a coordinate;
a position detection step of detecting a position of the rotor;
a calculation step of calculating a control magnetic pole position or a magnetic pole deviation amount based on at least one of the rotor position and a position command and the rotor speed and a speed command;
an acceleration calculation step of calculating an acceleration of the rotor;
Equipped with
a magnetic pole position of the rotor is estimated by performing coordinate transformation in the feedback-side coordinate transformation step and the command-side coordinate transformation step based on the control magnetic pole position or the magnetic pole deviation amount calculated in the calculation step;
The current command includes a dither signal.
The magnitude of the dither signal is controlled based on the acceleration of the rotor calculated in the acceleration calculation step or the speed of the rotor.

A magnetic pole position estimation method for estimating a magnetic pole position of a rotor of a synchronous motor according to one aspect of the present invention includes the steps of:
a first step of performing current control based on a q-axis current command to converge a controlled magnetic pole position or a magnetic pole deviation amount to a first convergence value;
a second step of performing current control based on a d-axis current command to cause the controlled magnetic pole position or the magnetic pole deviation amount to converge to a second convergence value;
a third step of performing current control based on a current command obtained by adding a dither signal to the q-axis current command or the d-axis current command, and converging the controlled magnetic pole position or the magnetic pole deviation amount to a third convergence value;
Equipped with
The first step, the second step, and the third step are
a current detection step of detecting a current flowing through the synchronous motor;
a feedback side coordinate transformation step of transforming a current detection value detected in the current detection step into a coordinate transformation state;
a current control step of calculating a voltage command from the current command and the current feedback that has been coordinate-converted in the feedback-side coordinate conversion step;
a command-side coordinate conversion step of converting the voltage command or the current command into a coordinate;
a position detection step of detecting a position of the rotor;
a calculation step of calculating the control magnetic pole position or the magnetic pole deviation amount based on at least one of the rotor position and a position command and the rotor speed and a speed command;
having
The third step comprises:
The method further includes an acceleration calculation step of calculating an acceleration of the rotor,
In the third step, a magnitude of the dither signal is controlled based on the acceleration of the rotor calculated in the acceleration calculation step or the speed of the rotor.

According to the present invention, even when a synchronous motor is used in a mechanical system in which static friction or Coulomb friction exists, it is possible to provide a magnetic pole position estimation device that reduces the shock and vibration that is applied to the mechanical system and can estimate the rotor's magnetic pole position with high accuracy using a small number of steps.

1 is a block diagram showing a configuration of a magnetic pole position estimation device according to an embodiment of the present invention. 4 is a flowchart for estimating a magnetic pole position of a rotor by the magnetic pole position estimating device according to the embodiment of the present invention. 4 shows a simulation result of the magnetic pole position estimation device according to the embodiment of the present invention. 13 shows a simulation result of the magnetic pole position estimation device according to the reference example.

(Embodiments of the present invention)
Hereinafter, an embodiment of the present invention will be described with reference to the drawings. For the sake of convenience, the description of the members having the same reference numbers as those already described in the description of the embodiment will be omitted.

FIG. 1 is a block diagram showing a configuration of a magnetic pole position estimation device according to an embodiment of the present invention.
As shown in Fig. 1, the control device for the servo motor M (hereinafter also referred to as a magnetic pole position estimation device) is composed of a dq-axis current command setter 1, a q-axis soft start controller 2, a d-axis soft start controller 3 (hereinafter, the q-axis soft start controller 2 and the d-axis soft start controller 3 are also collectively referred to as a soft start controller), a q-axis current controller 4, a d-axis current controller 5 (hereinafter, the q-axis current controller 4 and the d-axis current controller 5 are also collectively referred to as a current control unit), a command side coordinate converter 6, a PWM controller 7, a power converter 8, a current detector 9, a feedback side coordinate converter 10, a position detector (encoder) 11, a magnetic pole position calculator 12, a first filter 13, a magnetic pole deviation amount storage unit 14, an acceleration calculator 15, a second filter 16, a magnitude controller 17, and a dither signal generator 18. The magnetic pole position calculator 12 includes a speed calculator 19, a position controller 20, and a speed controller 21. The dq-axis current command setter 1 and the magnitude controller 17 are controlled by a magnetic pole position estimation controller 30. The position controller 20 is constituted by, for example, a proportional controller, and the speed controller 21 is constituted by, for example, a proportional-integral controller.

The dq-axis current command setter 1 outputs a q-axis current command set value. The q-axis soft start controller 2 performs soft start control of the q-axis current command set value input from the dq-axis current command setter 1. Here, the soft start control means gradually increasing the current command set value from the initial value to a predetermined value at a constant time rate. The dither signal generator 18 generates and outputs a dither signal. Here, the dither signal is a waveform that alternately oscillates in the positive direction and the negative direction, such as a triangular wave or a sine wave, and has the same value of the amplitude in the positive direction and the amplitude in the negative direction, and is a value that generates a torque larger than static friction or Coulomb friction. The dither signal is added to the soft start controlled q-axis current command set value to calculate the q-axis current command IqC. The deviation between the calculated q-axis current command IqC and the q-axis current feedback IqF from the feedback side coordinate converter 10 is input to the q-axis current controller 4. The q-axis current controller 4 calculates the q-axis voltage command VqC from the input deviation.

The dq-axis current command setter 1 outputs a d-axis current command set value. The d-axis soft-start controller 3 performs soft-start control on the d-axis current command set value input from the dq-axis current command setter 1, and outputs a d-axis current command IdC. The deviation between the output d-axis current command IdC and the d-axis current feedback IdF from the feedback side coordinate converter 10 is input to the d-axis current controller 5. The d-axis current controller 5 calculates the d-axis voltage command VdC from the input deviation.

The calculated q-axis voltage command VqC and d-axis voltage command VdC are input to the command-side coordinate converter 6. The command-side coordinate converter 6 converts the q-axis voltage command VqC and the d-axis voltage command VdC from the dq rotating coordinate system to the three-phase stationary coordinate system based on the control magnetic pole position θc, and calculates the three-phase voltage commands VUC, VVC, and VWC. The control magnetic pole position θc is calculated by adding the rotor position θFB to the magnetic pole deviation amount estimated value θe output from the magnetic pole position calculation unit 12. The calculated three-phase voltage commands VUC, VVC, and VWC are input to the PWM controller 7, which generates and outputs PWM control signals from the input three-phase voltage commands VUC, VVC, and VWC. The output PWM control signal is input to the power converter 8, and the power converter 8 drives the servo motor M based on the PWM control signal.

In this embodiment, the q-axis voltage command VqC and the d-axis voltage command VdC are input to the command-side coordinate converter 6 and are coordinate-converted from the dq rotating coordinate system to the three-phase stationary coordinate system. However, the q-axis current command IqC and the d-axis current command IdC may be input to the command-side coordinate converter 6 and are coordinate-converted from the dq rotating coordinate system to the three-phase stationary coordinate system.

The current detector 9 detects the motor current IU of the U phase of the servo motor M and the motor current IV of the V phase. The current detection values of the motor currents IU and IV are input to the feedback side coordinate converter 10. The feedback side coordinate converter 10 converts the current detection values of the motor currents IU and IV from a three-phase stationary coordinate system to a dq rotating coordinate system based on the control magnetic pole position θc, and calculates the q-axis current feedback IqF and the d-axis current feedback IdF.

The position detection unit (encoder) 11 is mounted on the servo motor M and detects the rotor position θFB of the servo motor M. The detected rotor position θFB is input to a speed calculator 19 and an acceleration calculator 15. The speed calculator 19 differentiates the detected rotor position θFB to calculate and output the rotor speed VFB. The acceleration calculator 15 differentiates the detected rotor position θFB twice to calculate and output the rotor acceleration AFB.

The deviation between the position command θcmd and the detected rotor position θFB is input to the position controller 20, which outputs the speed command Vcmd. The deviation between the output speed command Vcmd and the rotor speed VFB is input to the speed controller 21, which calculates the magnetic pole deviation estimation value θe. The rotor position θFB is added to the calculated magnetic pole deviation estimation value θe to calculate the control magnetic pole position θc. The calculated control magnetic pole position θc is input to the command side coordinate converter 6 and the feedback side coordinate converter 10. The calculated magnetic pole deviation estimation value θe is input to the first filter 13. The first filter 13 is composed of a low-pass filter or the like, and suppresses fluctuations in the magnetic pole deviation estimation value θe due to the dither signal. The magnetic pole deviation estimation value θe that has passed through the first filter 13 is stored in the magnetic pole deviation storage device 14.

The rotor acceleration AFB output from the acceleration calculator 15 is input to the second filter 16. The second filter 16 is composed of a low-pass filter or the like, and averages the dither frequency components of the rotor acceleration AFB to suppress fluctuations due to the dither signal. The rotor acceleration AFB that has passed through the second filter 16 is input to the magnitude controller 17. The magnitude controller 17 generates and outputs a magnitude signal S for controlling the magnitude of the dither signal based on the input rotor acceleration AFB. The output magnitude signal S is input to the dither signal generator 18, which generates a dither signal with a magnitude (amplitude) based on the magnitude signal S.

In the above configuration, the position command θcmd is set to 0, and the speed controller 21 is configured as a proportional-integral controller, so that the steady-state deviation between the speed command Vcmd and the rotor speed VFB is controlled to 0. As a result, the detected rotor position θFB ultimately converges to 0.

In this embodiment, the position command θcmd is input, but the speed command Vcmd may be input without providing the position controller 20.

FIG. 2 is a flowchart showing a process for estimating the magnetic pole position of the rotor by the magnetic pole position estimating device according to the embodiment of the present invention.
When the magnetic pole position estimation device starts the magnetic pole position estimation flow, in the first step, the q-axis current command setting value of the dq-axis current command setter 1 is set to a constant value, the d-axis current command setting value is set to 0, the magnitude signal S of the magnitude controller 17 is set to 0, and the synchronous motor is driven (S101).

As a result, the q-axis current command IqC is gradually increased at a constant time rate by the q-axis soft-start controller 2, and torque is generated gradually in the rotor according to the magnetic pole position error, which is the deviation between the magnetic pole position θre of the rotor and the control magnetic pole position θc, and the rotor position θFB moves due to the torque. Then, a deviation occurs between the position command θcmd and the detected rotor position θFB, and a speed command Vcmd is output from the position controller 20. Furthermore, a deviation occurs between the speed command Vcmd and the rotor speed VFB, and the magnetic pole deviation amount estimate value θe is calculated by the speed controller 21. The calculated magnetic pole deviation amount estimate value θe is added to the rotor position θFB to calculate the control magnetic pole position θc. The control magnetic pole position θc is input to the command side coordinate converter 6 and the feedback side coordinate converter 10, and the current control of the servo motor M is performed. Therefore, since only the q-axis current is flowing, the magnetic pole position error converges to near π/2 (first convergence value), and the torque generated in the rotor after convergence becomes almost 0. Because the control system operates in a direction that suppresses the position deviation, the detected rotor position θFB returns to its original position from the position to which it moved. Note that if the initial value of the magnetic pole position error is -π/2, the torque generated in the rotor will be 0, the detected rotor position θFB will not move, and the magnetic pole position error will remain at -π/2.

After a certain time has elapsed in the first step until the rotor position θFB returns to its original position, in the second step, the q-axis current command setting value of the dq-axis current command setting device 1 is set to 0, the d-axis current command setting value is set to a constant value, the magnitude signal S of the magnitude controller 17 is set to 0, and the synchronous motor is driven (S102).

As a result, the d-axis current command IdC is gradually increased at a constant time rate by the d-axis soft-start controller 3, and torque is generated gently in the rotor according to the magnetic pole position error, and the rotor position θFB moves due to the torque. Then, as in the first step, the speed controller 21 calculates the magnetic pole deviation amount estimate θe. The calculated magnetic pole deviation amount estimate θe is added to the rotor position θFB to calculate the control magnetic pole position θc. The calculated control magnetic pole position θc is input to the command side coordinate converter 6 and the feedback side coordinate converter 10, and the current control of the servo motor M is performed. Therefore, the torque generated in the rotor is controlled to be small, the magnetic pole position error converges to 0 (second convergence value), and the torque generated in the rotor after convergence becomes almost 0.

After a certain time has elapsed until the rotor position θFB returns to its original position in the second step, in the third step, the initial value of the magnitude signal S is set to a small value, a dither signal such as an M-series signal or a triangular wave that alternately oscillates in the positive and negative directions is generated by the dither signal generator 18, and the magnitude of the dither signal is automatically adjusted by the magnitude controller 17 (S102). Specifically, the rotor acceleration AFB calculated by the acceleration calculator 15 passes through a second filter 16 that averages and detects the dither frequency components, and is input to the magnitude controller 17. The magnitude controller 17 compares the rotor acceleration AFB after passing through the second filter 16 with a reference value, and if the rotor acceleration AFB is smaller than the reference value, the magnitude signal S is gradually increased. If the rotor acceleration AFB is larger than the reference value, the magnitude signal S is left unchanged. The reference value is set to a value as small as possible within the range in which the motor can move beyond friction, so that the rotor movement due to the dither signal is minimized.

As a result, the dither signal gradually increases until the rotor acceleration AFB reaches the reference value, and the magnitude of the dither signal is automatically adjusted until the motor torque output by the dither signal slightly exceeds the friction and the motor can move. In this way, by gradually increasing the dither signal, vibrations due to the dither signal are prevented from being applied to the mechanical system, and the motor is only moved to a level that slightly exceeds the friction, so that the rotor magnetic pole position θre can be estimated while minimizing the effect on the mechanical system. Furthermore, by automatically adjusting the magnitude of the dither signal, it is possible to accommodate various magnitudes of friction torque.

The frequency of the dither signal is set to a high frequency, for example 500 Hz to 1000 Hz, so that vibrations due to the dither signal are not transmitted to the load side of the power transmission mechanism at frequencies higher than the anti-resonance frequency of the mechanical power transmission mechanism. When the motor side is moving, the rotor magnetic pole position θre can be estimated without being affected by friction, so there is no problem even if the load side of the power transmission mechanism does not move. In addition, when using a signal with a constant frequency such as a triangular wave, by setting the frequency different from the anti-resonance frequency or resonance frequency of the mechanical system, it is possible to avoid situations where the dither signal is not effective due to the anti-resonance frequency or where the dither signal excites the resonance frequency.

After a certain time has elapsed until the magnetic pole deviation estimate θe converges in the third step, the magnetic pole deviation estimate θeF that has passed through the first filter 13 is stored in the magnetic pole deviation storage 14, and the magnetic pole position estimation flow ends. After the magnetic pole position estimation flow ends, the magnetic pole deviation estimate θeF stored in the magnetic pole deviation storage 14 is added to the detected rotor position θFB to calculate the control magnetic pole position θc, making it possible to drive the synchronous motor at an accurate rotor magnetic pole position.

As described above, the magnetic pole position estimation method according to an embodiment of the present invention reduces the shock and vibration to the mechanical system even when a synchronous motor is used in a mechanical system where static friction or Coulomb friction exists, and enables highly accurate estimation of the rotor magnetic pole position with a small number of steps.

Figure 3 shows the results of a simulation of a magnetic pole position estimation device according to an embodiment of the present invention. The horizontal axis in (A) to (G) of Figure 3 indicates time t, in seconds (s). When 0s ≤ t < 0.1s, the first step (S101) shown in Figure 2 is performed, when 0.1s ≤ t < 0.2s, the second step (S102) shown in Figure 2 is performed, and when 0.2s ≤ t, the third step (S103) shown in Figure 2 is performed.

Figure 3 (A) shows the waveforms of the q-axis current command IqC and the q-axis current feedback IqF. Figure 3 (B) shows the waveforms of the d-axis current command IdC and the d-axis current feedback IdF. In the first step (0s ≤ t < 0.1s), the q-axis current command setting value of the dq-axis current command setting device 1 is set to a constant value, the d-axis current command setting value is set to 0, and the magnitude signal S of the magnitude controller 17 is set to 0. Therefore, the q-axis current command IqC is gradually increased at a constant time rate to a constant value of 3A by the q-axis soft start controller 2. The d-axis current command IdC becomes 0A. In the second step (0.1s ≤ t < 0.2s), the q-axis current command setting value of the dq-axis current command setting device 1 is set to 0, the d-axis current command setting value is set to a constant value, and the magnitude signal S of the magnitude controller 17 is set to 0. Therefore, the q-axis current command IqC becomes 0A. The d-axis current command IdC is gradually increased at a constant time rate to a constant value of 3A by the d-axis soft start controller 3. In the third step (0.2s≦t), the initial value of the magnitude signal S is set to a reference value, and the dither signal generator 18 generates a dither signal of an M-sequence signal that oscillates alternately in the positive and negative directions. Therefore, the q-axis current command IqC has a waveform that oscillates around 0A due to the dither signal, and the d-axis current command IdC remains at 3A. Note that the q-axis current feedback IqF and the d-axis current feedback IdF have waveforms that follow the q-axis current command IqC and the d-axis current command IdC, respectively.

Fig. 3C shows the waveform of the torque generated in the rotor, Fig. 3D shows the waveform of the rotor acceleration AFB, and Fig. 3E shows the waveform of the magnitude signal S. At the start of the first process (t = 0s), the q-axis soft-start controller 2 performs soft-start control, so the q-axis current command IqC gradually increases, and torque is generated slowly as shown in Fig. 3C. The slowly generated torque also moves the rotor position θFB, and a slowly small rotor acceleration AFB is generated as shown in Fig. 3D. After that, the magnetic pole position error converges to near π/2, and the torque generated in the rotor after convergence becomes almost 0 (a value equivalent to friction).
At the start of the second step (t=0.1 s), the d-axis soft-start controller 3 performs soft-start control, so the d-axis current command IdC gradually increases, and a gentle torque is generated as shown in Fig. 3C. The gentle torque causes the rotor position θFB to move, and a gentle, small rotor acceleration AFB is generated as shown in Fig. 3D. Thereafter, the magnetic pole position error converges to 0, and the torque generated in the rotor after convergence becomes almost 0 (a value equivalent to friction).
In the third step (0.2s≦t), as shown in FIG. 3E, the magnitude controller 17 gradually increases the magnitude signal S, and automatically adjusts the magnitude of the dither signal until the motor torque slightly exceeds the friction and the servo motor M can operate. As a result, as shown in FIG. 3C and FIG. 3D, the magnitude of the torque generated in the rotor and the rotor acceleration AFB are suppressed to small values.

Figure 3 (F) shows the waveform of the rotor position θFB. Figure 3 (G) shows the waveform of the magnetic pole position error. After a certain amount of time has elapsed from the start of the third process until the control magnetic pole position θc converges (0.4s≦t), the magnetic pole position error converges to 0 due to the dither signal. This makes it possible to estimate the rotor magnetic pole position with high accuracy.

As a result, the dither signal gradually increases until the rotor acceleration AFB reaches the reference value, and the magnitude of the dither signal is automatically adjusted until the motor torque output by the dither signal slightly exceeds friction and the motor can move. By gradually increasing the dither signal, sudden acceleration applied to the mechanical system is prevented, and the motor is only moved to a level that slightly exceeds friction, making it possible to estimate the rotor magnetic pole position θre with minimal impact on the mechanical system. In addition, the magnitude of the dither signal can be automatically adjusted to accommodate various magnitudes of friction torque.

(Reference example)
FIG. 4 shows a simulation result of the magnetic pole position estimation device according to the reference example when Coulomb friction exists. In the magnetic pole position estimation device according to the reference example, unlike the magnetic pole position estimation device according to the embodiment of the present invention, the soft start control by the soft start controller is not performed, and the magnitude controller 17 is not automatically adjusted to adjust the magnitude of the dither signal, and the magnetic pole position estimation operation is performed on a mechanical system with Coulomb friction. (A) to (G) in FIG. 4 show the same waveforms as (A) to (G) in FIG. 3. The horizontal axis of (A) to (G) in FIG. 4 indicates time t, and the unit is seconds (s). When 0≦t<0.1, the first step (S101) shown in FIG. 2 is performed, when 0.1≦t<0.2, the second step (S102) shown in FIG. 2 is performed, and when 0.2≦t, the third step (S103) shown in FIG. 2 is performed.

At the start of the first process (t = 0s), the q-axis soft start controller 2 does not perform soft start control, so the q-axis current command IqC is applied in steps, and a large torque is generated as shown in (C) of FIG. 4. At the start of the second process (t = 0.1s), the d-axis soft start controller 3 does not perform soft start control, so the d-axis current command IdC is applied in steps, and a large torque is generated as shown in (C) of FIG. 4. Furthermore, in the third process (0.2s ≦ t), the initial value of the magnitude signal S is set to a reference value. The dither signal needs to be set to a large value (amplitude) taking into account various magnitudes of Coulomb friction torque, and the reference value of the magnitude signal S also needs to be set to a large value. Therefore, as shown in (C) and (D) of FIG. 4, the torque and rotor acceleration AFB when the dither signal is generated are larger than the torque and rotor acceleration AFB in the embodiment of the present invention shown in (C) and (D) of FIG. 3. Therefore, a large torque is generated in the rotor, and the impact and vibration on the mechanical system become large.

As described above, the magnetic pole position estimation device according to an embodiment of the present invention is capable of estimating the rotor magnetic pole position with high accuracy and with a small number of steps, even when a synchronous motor is used in a mechanical system in which static friction or Coulomb friction is present, by including a dither signal in the current command or adding a dither signal to the control magnetic pole position.

Although an embodiment of the present invention has been described above, it goes without saying that the technical scope of the present invention should not be interpreted as being limited by the description of this embodiment. This embodiment is merely an example, and it will be understood by those skilled in the art that various modifications of the embodiment are possible within the scope of the invention described in the claims. The technical scope of the present invention should be determined based on the scope of the invention described in the claims and its equivalents.

In the embodiment of the present invention, the automatic adjustment of the dither signal magnitude is performed based on the calculated acceleration, but it may also be performed based on the calculated speed. In that case, however, the amount of movement of the mechanical system will be large.

1: dq axis current command setter 2: q axis soft start controller 3: d axis soft start controller 4: q axis current controller 5: d axis current controller 6: command side coordinate converter 7: PWM controller 8: power converter 9: current detector 10: feedback side coordinate converter 11: position detector (encoder)
12: Magnetic pole position calculation unit 13: First filter 14: Magnetic pole deviation amount storage unit 15: Acceleration calculator 16: Second filter 17: Magnitude controller 18: Dither signal generator M: Servo motor S: Magnitude signal

Claims (7)

A magnetic pole position estimation device for estimating a magnetic pole position of a rotor of a synchronous motor, comprising:
a current detector for detecting a current flowing through the synchronous motor;
a feedback side coordinate converter that converts a current detection value detected by the current detector into a coordinate converter;
a current control unit that calculates a voltage command from a current command and the current feedback that has been coordinate-converted by the feedback-side coordinate converter;
a command-side coordinate converter that converts the voltage command or the current command into a coordinate;
a position detection unit for detecting a position of the rotor;
a magnetic pole position calculation unit that calculates a control magnetic pole position or a magnetic pole deviation amount based on at least one of the rotor position and a position command and the rotor speed and a speed command;
an acceleration calculator for calculating an acceleration of the rotor;
Equipped with
the feedback-side coordinate converter and the command-side coordinate converter perform coordinate conversion based on the control magnetic pole position or the magnetic pole deviation amount calculated by the magnetic pole position calculation unit, so that the magnetic pole position estimation device estimates the magnetic pole position of the rotor;
The current command includes a dither signal.
A magnetic pole position estimation device, characterized in that a magnitude of the dither signal is controlled based on the acceleration of the rotor calculated by the acceleration calculator or the speed of the rotor. The magnetic pole position estimation device according to claim 1, which controls the current command with a soft start to gradually increase the current command value. controlling the dither signal to gradually increase its magnitude when the acceleration or speed of the rotor is less than a reference value;
2. The magnetic pole position estimation device according to claim 1, wherein when the acceleration or speed of the rotor is equal to or greater than a reference value, the magnitude of the dither signal is controlled so as to be a constant value. The magnetic pole position calculation unit
2. The magnetic pole position estimation device according to claim 1, further comprising: a position controller that calculates the control magnetic pole position from a deviation between the rotor position and the position command; or a speed controller that calculates the magnetic pole deviation amount from a deviation between the rotor speed and the speed command. The magnetic pole position calculation unit
5. The magnetic pole position estimation device according to claim 1, wherein the control magnetic pole position is calculated by adding the position of the rotor to the magnetic pole deviation. A magnetic pole position estimation method for estimating a magnetic pole position of a rotor of a synchronous motor, comprising the steps of:
a current detection step of detecting a current flowing through the synchronous motor;
a feedback side coordinate transformation step of transforming a current detection value detected in the current detection step into a coordinate transformation state;
a current control step of calculating a voltage command from the current command and the current feedback that has been coordinate-converted in the feedback-side coordinate conversion step;
a command-side coordinate conversion step of converting the voltage command or the current command into a coordinate;
a position detection step of detecting a position of the rotor;
a calculation step of calculating a control magnetic pole position or a magnetic pole deviation amount based on at least one of the rotor position and a position command and the rotor speed and a speed command;
an acceleration calculation step of calculating an acceleration of the rotor;
Equipped with
a magnetic pole position of the rotor is estimated by performing coordinate transformation in the feedback-side coordinate transformation step and the command-side coordinate transformation step based on the control magnetic pole position or the magnetic pole deviation amount calculated in the calculation step;
The current command includes a dither signal.
A magnetic pole position estimation method, comprising: controlling a magnitude of the dither signal based on the acceleration of the rotor calculated in the acceleration calculation step or the speed of the rotor. A magnetic pole position estimation method for estimating a magnetic pole position of a rotor of a synchronous motor, comprising the steps of:
a first step of performing current control based on a q-axis current command to converge a controlled magnetic pole position or a magnetic pole deviation amount to a first convergence value;
a second step of performing current control based on a d-axis current command to cause the controlled magnetic pole position or the magnetic pole deviation amount to converge to a second convergence value;
a third step of performing current control based on a current command obtained by adding a dither signal to the q-axis current command or the d-axis current command, and converging the controlled magnetic pole position or the magnetic pole deviation amount to a third convergence value;
Equipped with
The first step, the second step, and the third step are
a current detection step of detecting a current flowing through the synchronous motor;
a feedback side coordinate transformation step of transforming a current detection value detected in the current detection step into a coordinate transformation state;
a current control step of calculating a voltage command from the current command and the current feedback that has been coordinate-converted in the feedback-side coordinate conversion step;
a command-side coordinate conversion step of converting the voltage command or the current command into a coordinate;
a position detection step of detecting a position of the rotor;
a calculation step of calculating the control magnetic pole position or the magnetic pole deviation amount based on at least one of the rotor position and a position command and the rotor speed and a speed command;
having
The third step comprises:
The method further includes an acceleration calculation step of calculating an acceleration of the rotor,
A magnetic pole position estimation method, characterized in that in the third step, a magnitude of the dither signal is controlled based on the acceleration of the rotor calculated in the acceleration calculation step or the speed of the rotor.

Images