JP5753474B2 - Synchronous motor controller - Google Patents

Description

Embodiments described herein relate generally to a synchronous motor control device.

An electric motor that uses field windings, permanent magnets, and iron cores for the rotor is called a synchronous motor because it energizes the stator in synchronization with the rotation of the electric motor and generates torque. More specifically, a magnetic flux is generated by using a field winding and torque is obtained by simply using a synchronous motor. A permanent magnet is used by a permanent magnet for the rotor, and a magnet is obtained by obtaining permanent magnet. A synchronous motor or a motor that obtains reluctance torque using only an iron core as a rotor is called a reluctance motor.

  The synchronous motor as described above is generally controlled to obtain a high torque with high efficiency by measuring a rotation angle during operation and energizing an appropriate phase angle.

  In order to measure the rotation angle, a rotation angle sensor represented by a resolver, an encoder (pulse generator) or the like is generally installed. For example, a resolver is rotated by observing the action of a rotor acting as a rotary transformer with a stator winding, with a rotor installed on the rotating shaft of the synchronous motor, applying high-frequency excitation from a stator installed on the stator side of the synchronous motor. Detect the angle. The encoder also installs a disk with a large number of slits on the axis of rotation, applies light or laser to the slits and observes the transmitted light, obtains pulse information according to the slits of the disk, and counts the pulses. To obtain the relative rotation angle.

  In the synchronous motor control system using the rotation angle sensor as described above, the positioning of the rotation angle sensor is an important factor that determines the drive performance. That is, for example, in the case of a permanent magnet synchronous motor, torque can be efficiently generated by flowing current in a direction that is 90 degrees electrically advanced with the magnet magnetic flux direction of the rotor as the reference direction of rotation. If there is an error in the reference position of the angle sensor, current cannot flow accurately in a direction advanced 90 degrees with respect to the magnet magnetic flux direction, and an error will always occur in the current phase as much as the reference position error. Become. If there is an error in the current phase, the generated torque decreases, leading to an increase in loss and a decrease in acceleration / deceleration performance. In the worst case, if the error is large, the sign of the generated torque is reversed, and it can be considered that the torque rotates in an unintended rotation direction.

  The positioning of the rotation angle sensor requires an operation for matching the stator / rotor of the sensor with the reference point of the stator / rotor of the synchronous motor. However, since it is difficult to install the sensor with high precision when the sensor is assembled to the synchronous motor due to restrictions in the assembly process or the like, adjustment is often performed on the motor control device side after assembly. Specifically, the rotation angle information output from the sensor is converted into a digital value by a resolver / digital converter or the like, and a value that compensates for an error at the zero point of the rotation angle sensor is measured and controlled by means described below. By storing it in the device, the mounting error during assembly is compensated.

FIG. 12 shows the relationship between the rotation angle of the synchronous motor, the rotation angle of the sensor, and the angular difference between the reference points of the respective stators and rotors. The angle difference Δθ between the synchronous motor and the sensor is expressed by the sum of the difference Δθs between the respective stator reference points and the difference Δθr between the rotor reference points, as shown in Equation (1).

The positioning adjustment of the rotation angle sensor means that a value corresponding to Δθ is measured and stored as a parameter in the control device.

  As an adjustment method, for example, in the case of a permanent magnet synchronous motor, generally, when the magnet magnetic flux direction is taken as the reference direction of the rotor and the speed is controlled by the control device and rotated to a sufficiently high rotational speed, the synchronous motor is connected to the terminal of the synchronous motor. It is utilized that the counter electromotive voltage that appears is a 90-degree advance phase of the magnet magnetic flux. When performing this operation with the control device, adjust the angle difference parameter so that the voltage command vector phase is 90 degrees ahead of the rotation angle when the current command is set to zero after rotating to a constant rotational speed. do it.

  When the synchronous motor to be adjusted is connected to the load machine, the phase terminal of the synchronous motor may be opened, and the synchronous motor terminal voltage when rotating from the load machine at a constant rotational speed may be measured. Since the terminal voltage at this time is the back electromotive voltage itself, the phase can be measured and Δθ can be determined based on the correlation with the angle output from the rotation angle sensor.

JP 2010-166701 A

The method as described above makes it possible to adjust the angle difference between the synchronous motor and the rotation angle sensor. However, since the method described above has to rotate the synchronous motor, it is driven in combination with the control device for adjustment. In addition to the need for a process, dedicated equipment is required, which takes time. In addition, after the adjusted value is stored in the control device, it is necessary to manage the synchronous motor and the adjustment value as a pair, which takes time for management.

  Also, if a synchronous motor or rotation angle sensor fails during operation of a device incorporating a synchronous motor, the failed synchronous motor or sensor will be replaced, but the relationship between the synchronous motor and the rotation angle of the sensor is due to replacement work. Therefore, after the replacement, it is necessary to make a state where the synchronous motor can be rotated independently. Alternatively, the synchronous motor main body must be removed and installed in a dedicated adjustment facility, and an adjustment test must be performed. In either case, work time and labor are required.

The present invention has been made to solve the above-described problems, and can adjust the angle difference between the synchronous motor and the rotation angle sensor without rotating the synchronous motor to a rotational speed at which a counter electromotive voltage can be observed for adjustment. An object is to provide an apparatus.

The synchronous motor control device of the embodiment is connected to a synchronous motor that has electric saliency in the rotor and generates torque by flowing an alternating current in a phase synchronized with the rotation of the rotor, and is synchronized with the synchronous motor. A rotation angle sensor that detects the rotation angle of the motor, an inverter that is connected to the synchronous motor and receives a control command for controlling the synchronous motor, and outputs to the inverter based on orthogonal coordinates defined in an arbitrary rotation angle direction of the synchronous motor Based on a current control unit that outputs a control command value, a high-frequency voltage command generation unit that generates a command to apply a voltage to the control command value output from the current control unit, a control command value and a command to apply the voltage An angle difference calculation unit that calculates a rotation angle difference between the angle of the synchronous motor in the direction of the electric salient pole and the rotation angle detected by the rotation angle sensor, and a rotation calculated by the angle difference calculation unit. Based on the angular difference, in order to correspond to the actual rotational angle of the angle of rotation sensor synchronous motor, and a rotation angle detecting unit for correcting the rotational angle sensor.

The control system figure of the synchronous motor of 1st Embodiment. The figure which shows the coordinate definition in the vector control of a permanent magnet synchronous motor. The wave form diagram of a rotation high frequency alternating voltage. The rotation phase diagram of a rotation high frequency alternating voltage. The figure which shows the sampling point of a high frequency current. The control system figure of the synchronous motor of 2nd Embodiment. The flowchart which shows the control operation of 2nd Embodiment. Explanatory drawing of the current waveform and process of 2nd Embodiment. The figure which shows a high frequency current locus. The control system figure of the synchronous motor of 3rd Embodiment. The flowchart which shows the control operation of 3rd Embodiment. The flowchart which shows the control operation of 4th Embodiment. The control system figure of the synchronous motor of 4th Embodiment. The system diagram of LL arithmetic processing. The figure which shows the sampling point of the sensor rotation angle of 4th Embodiment. The figure which shows the rotation angle definition of an electric motor and a rotation angle sensor.

Hereinafter, a control unit for adjusting a rotation angle sensor of a synchronous motor according to an embodiment will be described with reference to the drawings.

(First embodiment)
The first embodiment will be described in detail with reference to FIGS. 1 to 5. FIG. 1 shows an inverter 1, a synchronous motor 2, a rotation angle sensor 3, a control unit 4, a high frequency voltage command generation unit 5, a current control unit 6, a rotation angle setting unit 7, a γδ / 3 phase conversion 8, a PWM processing unit 9, An angle difference calculation unit 10, a three-phase / γδ conversion 11, and a rotation angle detection unit 12 are shown.
The synchronous motor 2 includes a stator and a rotor. A current is passed through a plurality of stator windings installed in the stator to generate a stator magnetic flux, and a torque is generated by the interaction between the rotor magnetic flux and the stator magnetic flux. In particular, it has a characteristic that a desired torque can be obtained by passing a current in a phase synchronized with rotation. The electrical saliency is such that when the motor is modeled in a rotation orthogonal two-axis coordinate system (generally called a dq-axis coordinate system) synchronized with the rotation of the motor, the d-axis inductance Ld and the q-axis inductance Lq are different. Refers to a characteristic. In the case of a permanent magnet synchronous motor having electrical saliency, Ld <Lq is often satisfied, and this characteristic is sometimes called reverse saliency. In a permanent magnet synchronous motor having reverse saliency, reluctance torque resulting from the characteristic Ld <Lq can be used, and the total generated torque can be improved.

  The rotation angle sensor 3 is installed on each of the stator side and the rotor side of the synchronous motor 2, and measures the rotation angle of the synchronous motor 2 based on the electromagnetic action and the optical action as described above, and corresponds to the rotation angle. The rotation angle signal D is output. The rotation angle signal D is generally an analog value or a pulse having a predetermined resolution. In the control device, these signals are converted into digital signals so that they can be used for control calculations as input, and a rotation angle is obtained.

The control device that controls the synchronous motor 2 includes an inverter 1 that converts electric power supplied to the motor from direct current to alternating current, and a control unit 4 that outputs a switching command to each element of the inverter 1.

  In FIG. 1, the rotation angle setting unit 7 outputs a desired rotation angle θset and outputs it to the γδ / 3-phase conversion unit 8 and the 3-phase / γδ conversion unit 11. θset is always a constant value, and the value may be set arbitrarily, or the output angle of the rotation angle sensor 3 at the time of setting may be used. When the output value of the rotation angle sensor 3 is used, even if the synchronous motor 2 rotates after setting and the sensor output fluctuates, the rotation angle setting unit 7 operates to output a constant rotation angle.

The PWM processor 9 generally uses a triangular wave comparison PWM. By comparing the voltage command and the triangular wave carrier, an ON / OFF gate command for each switching element of the inverter is generated.

The current control unit 6 controls the current of the synchronous motor 2 using a control method such as vector control. The vector control will be described below using a permanent magnet synchronous machine using a permanent magnet as a rotor as an example.

First, as shown in FIG. 2, an arbitrary reference direction corresponding to the stator (for example, the center direction of the U-phase winding) is defined as an α axis, and a direction orthogonal thereto is defined as a β axis, and the rotation angle of the synchronous motor 2 A dq axis coordinate system rotated by θm is defined. Similarly, a γδ coordinate system rotated by θset from the α axis is defined. Based on such a definition, the voltage / current relationship of the permanent magnet synchronous motor is expressed by the number of voltage steps of number (2).

As for the current command, the γ-axis current command I _γ ^ref and the δ-axis current command I _δ ^ref are given as shown in ^Equation (3) based on the torque command or the like.

Next, the current commands I _γ ^ref and I _δ ^ref, and the γ-axis response value I _γ ^res and the δ-axis response value I _δ ^res of the current flowing through the synchronous motor are input, for example, by proportional integral control as follows. , Γ-axis voltage command V _γ ^ref and δ-axis voltage command V _δ ^ref are calculated.

Next, the γ-axis voltage command V _γ ^ref and the δ-axis voltage command V _δ ^ref output as described above are converted into the following in the γδ / 3-phase conversion unit based on the rotation angle θset of the γδ coordinate system. Coordinate conversion is performed by calculation, and three-phase voltage commands V _u ^ref , V _v ^ref , V _w ^ref are calculated.

The three-phase voltage command obtained by the equation (5) is input to the PWM processing unit 9.

The current response used in the vector control is based on the rotation angle θset output from the rotation angle setting unit, and the three-phase / γδ conversion unit calculates the γ-axis current response value I _γ ^res , δ-axis current by the following calculation. A response value I _δ ^res is obtained.

The high frequency voltage command generation unit 5 generates a rotation high frequency alternating voltage command that rotates at a rotation angle θv that is a predetermined value. The high-frequency alternating voltage command is a voltage command that rotates on the γδ-axis coordinates at a rotation angle θv with a simple vibration voltage of the frequency fh as an amplitude, and is expressed as the following formulas (7) to (10).

FIG. 3 shows the high-frequency alternating voltage command determined as described above, and FIG. 4 shows θv at this time. 3 and 4, V = 50 [V], fh = 1000 [Hz], and fv = 10 [Hz].

The angle difference calculation unit 10 uses the rotational high-frequency current response flowing in the synchronous motor 2 by applying the rotational high-frequency alternating voltage command, and uses a half cycle of the rotational phase θv of the rotational high-frequency alternating voltage command, that is, a double angle of θv. The relative rotation angle θf of the rotor of the synchronous motor 2 as seen from the γδ coordinate system is calculated based on the coefficient obtained as a result of performing the first order Fourier series operation corresponding to one cycle.

Then, using the relative rotation angle θf, the rotation angle θset of the γδ coordinate system, and the rotation angle θs output by the rotation angle sensor 3, the angle difference Δθ between the rotation angle sensor 3 and the actual rotation angle as shown in Equation (11). Is output to the rotation angle detector 12.

  The rotation angle detection unit 12 obtains a rotation angle value based on the signal from the rotation angle sensor 3, adds the output Δθ from the angle difference calculation unit, and corrects the corrected value θc to the γδ / 3-phase conversion units 8 and 3. Output to the phase / γδ converter 11.

Next, the principle of calculating the relative rotation angle θf by the processing of the angle difference calculation unit 10 will be described. First, since the high frequency voltage is applied and the high frequency current is observed in the number of voltage steps (number (2)) of the permanent magnet synchronous motor, terms other than the differential term are sufficiently smaller than the differential term and can be ignored. Therefore, when the component related to the differential term is extracted, the number is (12).

When the number (12) is transformed into the γδ coordinate system and solved for the current, the number (13) is obtained.

In the equation (13), when the rotating high-frequency alternating voltage represented by the equations (7) to (10) is applied, a current response like the equation (14) is obtained.

When the component I _hv in the direction of the phase angle θv of the rotating high-frequency alternating voltage is extracted from the current response expressed by the number (14), the number (15) is obtained.

If the amplitude I _hvs of the sin θ _h component of I _hv is extracted from the number (15) and the cosine component a _θv and sine component b _θv related to 2θv are calculated by Fourier series expansion, the relative rotation angle θf is obtained as follows. be able to.

For the Fourier series expansion of the current, the general number (17) can be used.

In _order to extract the amplitude I _hvs of the sin θ _h component from I _hv , a simple method is to sample the current at the zero cross timing of the high-frequency voltage. The current sample timing is shown in FIG.

In FIG. 5, the amplitude I _hvs can be calculated from each sampling value as follows.

Since θf can be obtained by the above calculation, and θset and θs are known, the angle difference Δθ can be calculated by the equation (11).

Here, by setting the alternating frequency fh of the rotating high-frequency alternating voltage sufficiently high, the frequency of the high-frequency current flowing by this voltage is also increased. Since the torque generated by this current has positive and negative high frequencies, it can be said that the synchronous motor 2 hardly rotates by the generated torque during the above calculation.

In addition, the current detection at the timing of FIG. 5 does not necessarily have to be detected at a precisely designated time point, and if I _hvs can be obtained equivalently, a timing shift can be allowed. For example, by sampling at a plurality of points before and after each sampling point in FIG. 5 and taking an average over time, it is possible to obtain substantially the same value as when sampling was equivalently designated. In this case, the effect of being less susceptible to noise than a single sampling can be obtained.

Further, the amplitude Vh of the rotating high-frequency alternating voltage represented by the equation (7) may be a rectangular wave signal. Since the rectangular wave signal contains sinusoidal components of the same frequency and the same phase, it can be applied as it is without changing the principle, and the rectangular wave signal can be created simply by switching between positive and negative values in the control device. Calculation load can be reduced.

In this case, since the high-frequency current is expected to have a triangular wave shape, it is desirable to sample at each peak value of the triangular wave. Since this is equivalent to sampling at the positive / negative switching timing of the rectangular wave high-frequency alternating voltage, management of the sampling timing is also simplified.

Even if sampling is not performed at each peak of the triangular wave, an equivalent peak value can be derived by sampling at a sufficiently high speed, for example, and calculating the slope of the triangular wave.

(effect)
According to the control system of the control device of the embodiment described above, it is possible to easily correct the error of the rotation angle sensor without rotating the synchronous motor. For this reason, it is possible to reduce the adjustment man-hours when incorporating the rotation angle sensor, simplify the management of the synchronous motor and the control device, and simplify the replacement work in the event of a failure.

(Second Embodiment)
The second embodiment will be described in detail with reference to the drawings. In addition, about the thing which has the same structure as FIG. 1 thru | or 5, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  In this embodiment, a current current command setting unit 21 is added to the first embodiment. As shown in FIG. 6, the direct current command setting unit 21 sets current commands (Iγref and Iδref) in the γδ coordinate system and outputs them to the current control unit 6.

As shown in FIG. 7, the control system of the present embodiment includes a first step (S21) for applying a direct current command to fix the magnetic pole, confirmation of the stationary state of the synchronous motor (S22), and a rotating high-frequency alternating voltage. And a second step (S23) of correcting the rotation angle sensor with a high frequency current.

The purpose of the first step (S21) for applying the direct current command (S21) and the direct current command setting unit 21 is to fix the magnetic poles of the synchronous motor 2. When correcting the rotation angle sensor in the second step (S23), if the rotor of the synchronous motor 2 rotates while calculating the angle difference by the angle difference calculation unit 10, the angle difference of the calculation result Is not calculated correctly and an error occurs. Therefore, in the first step (S21), a direct current command is given to the angle of the γδ coordinate system that is rotated and fixed from the reference axis α at the phase of the rotation angle θset, and current control is performed. As a result, the magnetic poles of the synchronous motor 2 are fixed in the direction of the magnetic flux generated by the current. At that time, if the actual rotation angle of the synchronous motor 2 and the phase of the magnetic flux generated by the current from the reference axis α coincide, the rotor does not move. If they do not match, the rotor will rotate, but since the angle of rotation is less than 180 degrees in electrical angle, the mechanical angle will only move by a smaller angle.

FIG. 8 shows an example of current response when the control system of this embodiment is implemented. After direct current is passed in the first step (S21), the process proceeds to the second step (S23), and high-frequency current is flowing. In FIG. 8, a positive command value is given to the current command (Iγref), and a zero command value is given to the current command (Iδref). With this setting, the rotor is fixed in the direction of the γ axis of the γδ coordinate system. At this time, if a direct current is applied, the rotor may rotate and may move slightly oscillating with respect to the γ-axis, but if it is applied for a sufficiently long time, it attenuates and stops.

Furthermore, it is possible to shorten the time until the camera stops by performing speed control. In this case, the rotational speed of the rotor can be easily obtained by differentiating the rotational angle of the rotational angle sensor, and even if there is an error in the rotational angle, the speed is not affected, so an accurate speed can be obtained. Therefore, sufficient speed control is possible. For the speed control, the PI control generally used conventionally is sufficient, the input speed command is set to zero, the output torque command is given to the current command (Iδref), and the direct current is used as the current command (Iγref). The command is given as a constant value. With this configuration, if the synchronous motor 2 is not connected to the load, that is, a so-called free-run state, the current command (Iδref) when the rotor is stationary becomes zero, and the DC current command is applied in the Iγ direction. This is equivalent to the above state.

As described above, the control system of the present embodiment moves to the second step while the DC current command is fixed after it is stopped in the γ-axis direction by applying the DC current. The second step is the same as the processing in the first embodiment except that a direct current is passed.

Next, the effect of correcting the rotation angle by applying a direct current to make the rotor stationary will be described. It is desirable that the rotation angle correction process performed in the second step is a state in which the rotation angle of the rotor does not change during the correction calculation, that is, is stationary. Normally, if the rotor is not connected to the load, or is connected but the load is not operating, the rotor will not rotate unless torque is generated from the synchronous motor 2 side. However, since a high-frequency current flows during the correction calculation, instantaneous torque is generated. Since the high-frequency current is an alternating high-frequency current, the average torque becomes zero, but the possibility that the rotor rotates slightly by instantaneous torque is not zero. Therefore, it is possible to prevent the rotor from rotating due to the torque generated by the high-frequency current by applying a direct current and fixing the rotor during the correction calculation.

In addition, it is possible to eliminate an error in correction calculation due to the influence of the dead time of the inverter 1. A dead time period is provided for the gate signals of the upper and lower arms processed by the PWM processing unit 9 so that both the gate signals become L so that the upper and lower arms of the inverter are not short-circuited when the signal is switched between H and L. However, since the output voltage during the dead time period is determined by the direction of the phase current of the corresponding arm, a voltage different from the voltage command may be output. This situation can be avoided by performing dead time compensation that corrects different voltage commands during the dead time using the response value of the phase current, but this compensation is especially effective when the phase current is near zero. It will be difficult to do. In particular, when a high-frequency current is passed, a zero crossing of the phase current frequently occurs, so that an error also occurs in the angle difference calculation due to an output voltage error during the dead time period, and a correct correction value cannot be calculated.

The present embodiment in which a direct current is applied can avoid such a situation. In other words, by applying a direct current, it is possible to create a situation in which each phase current is always positive or negative, so the influence of the dead time on the output voltage is always constant regardless of the high-frequency current, so the correction calculation It becomes possible to eliminate the influence on the.

FIG. 9 shows a current locus schematically representing the above situation. If a current vector is included in any of the six regions surrounded by the dotted line in FIG. 9, the sign of the current of each phase does not switch between positive and negative, and it can be said that the influence of the dead time is uniform. By applying a direct current, even if a high-frequency current flows, the current sign of each phase is not switched . Even in any of the six regions surrounded by dashed line the same effect can be obtained. For example, before performing the first step (S21) of this embodiment, the second step (angle error correction calculation) (S23) is performed without applying a direct current, and some correction errors due to dead time are included. Even if it is, the approximate rotation angle is obtained, the region including the rotation angle is selected from the six regions, and the direct current is set so that the direct current flows through the central angle of the region. Thus, by performing the first step (S21) and then performing the second step (S23) again, it is possible to reliably eliminate the influence of the dead time.

By configuring as described above, the control unit of the present embodiment can accurately correct the error of the rotation angle sensor while preventing the rotation of the rotor during the correction calculation and eliminating the influence of the dead time. Therefore, it is possible to reduce the adjustment man-hours when incorporating the rotation angle sensor, simplify the management of the electric motor and the control device, and simplify the replacement work at the time of failure.

(Third embodiment)
The third embodiment will be described in detail with reference to FIGS. 10 to 11. In addition, about the thing which has the same structure as FIG. 1 thru | or 9, the same code | symbol is attached | subjected and description is abbreviate | omitted.

The control unit 30 of FIG. 10 is obtained by adding a DC current command setting unit 31, a magnetic pole discrimination current command setting unit 32, a magnetic pole polarity discrimination unit 33, and a changeover switch 34 to the first embodiment. The DC current command setting unit 31 sets current commands (Iγref and Iδref) in the γδ coordinate system and outputs them to the current control unit 6. The magnetic pole polarity determination current command setting unit 32 sets a current command for the magnetic pole polarity and outputs the current command to the current control unit 6. In the subsequent switch 34, it is selected which of the current command from the DC current command setting unit 31 and the current command from the magnetic pole pole discrimination current command setting unit 32 is connected to the current control unit 6. The magnetic pole polarity determination unit 33 determines whether or not the magnetic pole polarity is reversed based on the current response, and outputs a magnetic pole polarity determination command to the rotation angle detection unit 12.

A method for switching the changeover switch 34 will be described. In the DC current application process, the changeover switch 34 is switched so that the DC current command setting unit 31 and the current control unit 6 are connected, and the DC current command is input to the current control unit 6. On the other hand, during the magnetic pole polarity discrimination process, the changeover switch 34 is switched so that the magnetic pole polarity discrimination current command setting unit 32 and the current control unit 6 are connected, and the magnetic pole polarity discrimination current command is input to the current control unit 6.

In the angle difference calculation method according to the first embodiment, as apparent from the equation (16), the correction angle θf is obtained based on the double angle of the rotation phase of the high-frequency alternating voltage. On the other hand, an angle of 0 degrees or 180 degrees is calculated, and it is not known which is the angle. If the rotation angle sensor is installed at an accuracy of at least ± 90 degrees with respect to the actual magnetic pole position, the calculated correction angle is sure to be 0 degree, but the above accuracy is not satisfied. In this case, it is necessary to determine whether the angle is 0 degree or 180 degrees.

For the above reasons, in the present embodiment, the third step (S34) is added to the first step (S31), the step (S) 32 for confirming that the motor is stationary, and the second step (S33). . In the flowchart of FIG. 11, since the angle of the rotation angle sensor is corrected by the ambiguity of 0 degrees or 180 degrees until the second step (S32), the magnetic pole polarity is determined in the subsequent third step.

For example, in the case of a permanent magnet synchronous motor, a positive and negative DC current is applied in the magnet direction, and at the same time, a high-frequency alternating voltage is applied in the magnet direction in the third step (S34). When the magnetic flux generated by the direct current is in the same direction as the magnet magnetic flux, magnetic saturation occurs due to the sum of both magnetic fluxes, and the d-axis inductance decreases. In the reverse direction, both magnetic fluxes cancel each other, so that no magnetic saturation occurs and the d-axis inductance does not decrease. Inductance can be calculated by the amplitude of the high-frequency current that flows due to the applied high-frequency voltage, so that the polarity of the magnetic pole depends on the amplitude of the high-frequency current when a direct current is passed in the positive direction and when a direct current is passed in the negative direction Can be determined.

With the configuration as described above, the control unit of the present embodiment can eliminate the ambiguity of the correction angle due to the calculation principle and can accurately correct the error of the rotation angle sensor. It is possible to reduce the adjustment man-hours when incorporating the sensor, simplify the management of the motor and the control device, and simplify the replacement work at the time of failure.

Note that the methods described in the first to third embodiments can be used for estimating the rotation angle as they are, and the cost of the rotation angle sensor can be obtained by applying the method to a system in which the rotation angle sensor is not installed. As a result, it is possible to reduce the number of assembly steps and to construct a robust system free from problems caused by sensor disconnection.

(Fourth embodiment)
A fourth embodiment will be described with reference to FIGS. In addition, about the thing which has the same structure as FIG. 1 thru | or 9, the same code | symbol is attached | subjected and description is abbreviate | omitted.

In the present embodiment, an electric motor acceleration control process, a rotation speed determination process, and a rotation angle recorrection process shown in FIG. 12 are added. This will be described below.

(Function)
In FIG. 12, in addition to the 1st system 400 and the 2nd system 401, the process of the 3rd system 403 is added. Since the first system 400 and the second system 401 have been described in the first to third embodiments, they are omitted here.
In the motor acceleration control step (S44) of the third system 403, the synchronous motor 2 is accelerated using the corrected rotation angle obtained by correcting the rotation angle from the rotation angle sensor 3 in the rotation angle correction step (S43) of the first system 400. Control is performed. More specifically, as shown in FIG. 13, a torque command / current instruction converting unit 43 connected to the speed controller 41, the rotational speed calculating section 42 which is connected to the speed controller 41 and the rotation angle sensor 3 In addition, the rotation speed calculation unit 42 and the integration unit are connected, and the integration unit is connected to the γδ / 3-phase conversion 8 and the 3-phase / γδ conversion 11.
In this configuration, the sensor output rotation angle is input to the rotation speed calculation unit 42, and the rotation speed response obtained by performing differential calculation and the rotation speed command are input to the speed control unit 41. The speed control unit 41 calculates a torque command by PI control based on the rotation speed response and the rotation speed command. The calculated torque command is input to the torque command / current command conversion unit 43. The input torque command is converted into a current command, and is output to the current control unit 6 as a γδ axis current command. Finally, the gate command is output to the inverter 1 to control the speed of the synchronous motor 2. At this time, the rotation speed command value input to the speed control unit 41 is set to be equal to or higher than the determination speed in the subsequent rotation speed determination step (S45).

As an acceleration control method other than the motor acceleration control step (S44), there is a method of inputting a positive constant command value to I _δ ^ref . If I _δ current flows by current control according to I _δ ^ref in which a constant command value is input, a positive torque is generated. Since the magnetic pole polarity determination (S44) has already been performed in the second system 401, the synchronous motor 2 is accelerated in the positive rotation direction when a positive torque is generated.

At this time, if the load is connected to the synchronous motor 2 and acceleration is weak or not accelerated due to the load torque or the like, I _δ ^ref is increased or more appropriate as shown in the equation (3). What is necessary is just to change so that a current command may be given and a large acceleration torque may be obtained. In the case of this control method, since the speed is not controlled, it is necessary to finish the acceleration before reaching the excessive rotational speed. In this embodiment, since the rotation speed determination step (S45) is provided, the current command may be changed to zero when the determination speed is exceeded in this step. With this method, the speed control unit 41 is not necessary, and an effect that simplifies the control calculation can be obtained.

In the rotational speed determination step (S45), it is determined whether or not the rotational speed of the synchronous motor 2 has reached a predetermined value (determination speed, determination rotational speed) or more. The predetermined values (determination speed and determination rotation speed) are set to values (speed and rotation speed) such that the back electromotive voltage induced in the stator by the rotor magnetic flux of the synchronous motor 2 can be observed as a significant value. For example, if it is about 10% of the maximum rotation speed, it can be said that it is sufficient because there is an experimental result that can be observed as a significant value in a general permanent magnet synchronous motor. Naturally, the higher the voltage that can be observed, the less likely it is to be affected by various error factors, so it can be said that it is better to accelerate to the maximum allowable number of revolutions with 10% as the lower limit.

In the rotation angle recorrection step (S46), first, using the sensor output rotation angle θs of the rotation angle sensor 3, the current command value for controlling the current is set to zero, and the current flowing through the synchronous motor 2 is controlled to zero. Next, the voltage command phase angle θ _Vref corresponding to the control in which the current becomes zero is calculated. The voltage command phase angle θ _Vref is obtained by an arc tangent process as shown in Equation (19) using the current control output voltage shown in Equation (4).

When the current is zero, the voltage command phase angle θ _Vref is equal to the counter electromotive voltage caused by the rotor magnetic flux, and is exactly 90 degrees advanced from the rotor magnetic flux direction. _Therefore , 90 degrees is subtracted from the voltage command phase angle θ _Vref . The value is the rotor magnetic flux direction. Since the voltage command phase angle θ _Vref is a voltage command phase angle when controlled by the sensor output rotation angle θs, if 90 degrees is subtracted from this, the recorrection angle difference Δθ ₂ can be obtained immediately. Used to recorrect the sensor rotation angle. An equation for obtaining the re-correction angle difference Δθ ₂ is shown in Equation (20).

As a simpler method, there is a method in which a PLL (Phase Locked Loop) is configured so that the γ-axis voltage command V _γ ^ref becomes zero, and the rotation angle calculated by the PLL is used as the adjustment rotation angle. A block diagram of this calculation method is shown in FIG. FIG. 14 shows a proportional / integration processing unit 47, an integration unit 48, and a second integration unit 49 that are configured by a Kp calculation unit 47 a, a Ki calculation unit 47 b, and a first integration calculation unit 47 c. The γ-axis voltage command V _γ ^ref is input to the proportional / integral processing unit 47 and is multiplied by the values of Kp_PLL and Ki_PLL in the proportional / integral processing unit 47. The value multiplied by Ki_PLL is further integrated, and the two values are integrated by the integrating unit 48, and the adjustment speed ω _adj is calculated so that the γ-axis voltage command V _γ ^ref is zero. The adjustment speed ω _adj is input to the second integration calculation unit 49, and the adjustment rotation angle θ _{adj obtained} by integrating the adjustment speed ω _adj is output from the second integration calculation unit 49. As shown in FIG. 13, by using this adjustment rotation angle θ _adj as an input to the γδ / 3-phase conversion block 8 and the three-phase / γδ conversion block 11, the adjustment rotation angle θ _adj is used as a rotation angle. A loop is formed, and the adjustment rotation angle θ _adj at which the γ-axis voltage command V _γ ^ref becomes zero can be obtained.

Since the γ-axis voltage command V _γ ^ref becomes zero by the above control, the voltage command is only the δ-axis voltage command V _δ ^ref . For this reason, the δ-axis voltage command V _δ ^ref coincides with the counter electromotive voltage, but since the δ-axis is in a direction advanced by 90 degrees from the γ-axis, at this time, it is in the direction of the adjustment rotation angle θ _adj . The γ axis represents the rotor direction. Request re-correction angle difference [Delta] [theta] ₂ the rotation angle as the rotation angle θm of the synchronous motor 2. Note that the initial values of the adjustment speed ω _adj and the adjustment rotation angle θ _adj in the PLL may be the sensor rotation angular speed and the sensor output rotation angle θs at the start of this process. Here, it is assumed that an error is included in the sensor output rotation angle θs, but after the initial value is set, this error is corrected by the PLL, and the adjustment rotation angle θ _adj that does not include the error can be obtained.

Further, with reference to FIG. 15, a method that enables to shorten the processing time than the processing for determining the re-correction angle difference [Delta] [theta] ₂ as described above. The recorrection angle difference Δθ ₂ is calculated using the α-axis voltage V _α ^ref , the rotation angle θm, and the sensor output rotation angle θs shown in FIG. The α-axis voltage V _α ^{ref in} FIG. 15A is obtained by expressing the voltage command vector represented by the γ-axis voltage command V _γ ^ref and the δ-axis voltage command V _δ ^ref as a stationary coordinate αβ axis. This is a voltage command vector. The α-axis voltage V ^ref _α has a down zero cross timing 49 (timing for zero crossing from positive to negative). At the down zero cross timing 49, the sensor output rotation angle θs of FIG. How to re-correction angle difference [Delta] [theta] ₂ by using the rotational angle θm of the number (20) shown in FIG. 15 (b) calculated from the detected sensor output rotational angle [theta] s alpha-axis voltage V ^ref _alpha also. This method is based on the principle that the sensor rotation angle at this moment represents an angle error because the moment when the α-axis voltage that is the rotation angle reference axis zero-crosses from positive to negative is the moment when the rotation angle becomes zero. Yes. According to this method, control for managing the hold timing is required, but numerical calculation processing such as speed control and PLL becomes unnecessary, and the processing time can be shortened.

(effect)
In this embodiment, re-correction is performed based on the back electromotive voltage. However, in order to rotate to a sufficient speed at which the back electromotive force can be observed, it is necessary to generate a desired torque at least having a different sign, and for this purpose, there are a first rotation angle correction process and a magnetic pole determination process. If comprised in this way, the desired torque which does not differ in a code | symbol will be produced | generated, an electric motor can be rotated to sufficient speed, and also a rotation angle can be recorrected from the back electromotive voltage phase in that state. Therefore, highly accurate correction is possible.

With the configuration as described above, the control unit according to the present exemplary embodiment uses another method that is not affected by the current detection error even when the correction value itself includes an error due to the current detection error. By re-correction, it becomes possible to further improve the control accuracy.

All the embodiments described above are presented by way of example and do not limit the scope of the invention. Therefore, the present invention can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and modifications thereof are included in the scope of the invention described in the claims and equivalents thereof.

DESCRIPTION OF SYMBOLS 1 Inverter 2 Synchronous motor 3 Rotation angle sensor 4 Control part 5 High frequency voltage command generation part 6 Current control part 7 Rotation angle setting part 8 γδ / 3 phase conversion 9 PWM processing part 10 Angle difference calculation part 11 Three phase / γδ conversion 12 Rotation Angle detection unit 20 Control unit 21 DC current command setting unit 30 Control unit 31 DC current command setting unit 32 Magnetic pole determination current command setting unit 33 Magnetic pole polarity determination unit 34 Changeover switch 40 Control unit 41 Speed control unit 42 Rotational speed calculation unit 43 Torque command / current command conversion unit 44 proportional / integration processing unit 47 proportional / integration unit 47a Kp calculation unit 47b Ki calculation unit 47c first integration calculation unit 48 integration unit 49 second integration unit

Claims (4)

A synchronous motor having electrical saliency on the rotor and generating torque by flowing an alternating current in a phase synchronized with the rotation of the rotor;
A rotation angle sensor connected to the synchronous motor and detecting a rotation angle of the synchronous motor; an inverter connected to the synchronous motor and receiving a control command for controlling the synchronous motor;
Based on orthogonal coordinates defined in an arbitrary rotation angle direction of the synchronous motor, a current control unit that outputs a control command value to be output to the inverter;
A high frequency voltage command generation unit that generates a command to apply a voltage to the control command value output by the current control unit;
Based on the control command value and the command to apply the voltage, an angle difference calculation unit that calculates a rotation angle difference between the angle of the electric salient pole of the synchronous motor and the rotation angle detected by the rotation angle sensor;
Based on the rotation angle difference calculated by the angle difference calculation unit, a rotation angle detection unit that corrects the rotation angle sensor in order to associate the rotation angle sensor with the actual rotation angle of the synchronous motor;
A DC current is supplied to the synchronous motor to fix the rotor, and the inverter is driven by a rotating high-frequency alternating voltage command based on the control command value to which a voltage generated in the high-frequency voltage command generating unit is applied. An angle of the electric salient pole direction is obtained from the rotational high-frequency current response obtained from the first-order Fourier series calculation corresponding to one cycle of a double angle of the rotation phase of the voltage, and the angle difference calculation unit Magnetic pole position adjusting means for calculating a rotation angle difference;
A synchronous motor control device having: In the magnetic pole position adjusting means,
After an angular difference of electric salient pole direction orthogonal coordinates and synchronous motor by calculating the correction of the rotational angle sensor, the synchronous motor control device of claim 1, further comprising a pole polarity discriminating means for implementing the magnetic pole polarity discrimination. In the magnetic pole position adjusting means,
After correction of the rotational angle sensor, the synchronous motor is rotated until the rotational speed of the counter electromotive voltage can be sufficiently observed, further comprising a counter electromotive voltage correcting unit for re-correcting the rotational angle sensor on the basis of the counter electromotive voltage The synchronous motor control device according to claim 1 . In the magnetic pole position adjusting means ,
The direct current synchronous motor controller according to claim 1, wherein it has been characterized to be a value larger than the amplitude of the high-frequency current flowing through the rotation frequency alternating voltage.

Images