JP2022120508A - Motor control device - Google Patents

Abstract

To provide a motor control device capable of calculating the position of a rotor while reducing the cost.SOLUTION: A motor control device 1 includes a rotor position calculator 15 for calculating the position of a rotor 20 by using an estimated position estimated by a position estimator 9, and a rotation pulse. When the rotation of the motor 2 is started, the rotor position calculator 15 performs phase difference calculation processing S90 of calculating the phase difference between the position of the rotor 20 estimated by the position estimator 9 and the position based on the rotation pulse. This makes it possible to grasp where a reference position (d-axis) of the rotor 20 exists within a pitch angle range. Therefore, in the rotor position calculation process S110, the rotor position calculator 15 can accurately calculate an actual rotor position by performing calculation based on the phase difference in addition to the position of the rotor 20 based on the rotation pulse.SELECTED DRAWING: Figure 1

Description

The present invention relates to a motor control device.

2. Description of the Related Art As a conventional motor control device, a technique such as that described in Patent Document 1, for example, is known. A motor control device described in Patent Document 1 includes a motor including a stator and a rotor having a permanent magnet, a command section that transmits a command signal to the motor, and a position estimation section that estimates the position of the rotor. Prepare. This motor control device also includes a resolver for detecting the position of the rotor of the rotating motor. The motor control device performs current control after correcting the rotor position by estimating the error angle of the phase angle based on the disturbance observer using the output from the resolver.

JP 2014-158336 A

By the way, in the above conventional technology, since the resolver is used, it is necessary to adjust the reference position of the resolver and the position of the rotor before assembly. Also, the resolver requires a circuit for signal processing on the inverter side. As described above, the conventional motor control device is required to calculate the position of the rotor while reducing the cost.

SUMMARY OF THE INVENTION An object of the present invention is to provide a motor control device capable of calculating the position of a rotor while reducing costs.

A motor control device according to an aspect of the present invention includes a motor including a stator and a rotor having a permanent magnet, a command section that transmits a command signal to the motor, and a harmonic superimposition section that superimposes a harmonic on the command signal. a position estimator for estimating the position of the rotor by superimposing harmonics, a rotation pulse sensor for detecting the rotation pulse of the motor, the estimated position estimated by the position estimator, and the rotation pulse to determine the position of the rotor. a rotor position calculation unit that calculates a position, the command unit performs start processing for starting rotation of the motor in a state in which the initial position of the rotor is estimated, and the rotor position calculation unit is the position estimation unit. A phase difference calculation process is performed to calculate the phase difference between the rotor position estimated in and the rotor position based on the rotation pulse, and the rotor position is calculated based on the rotor position based on the rotation pulse and the phase difference Rotor position calculation processing for calculating the position is performed.

Such a motor control device includes a position estimator that estimates the position of the rotor by superimposing harmonics, and a rotation pulse sensor that detects rotation pulses of the motor. Here, if only the position estimated by the position estimator is used, it is not possible to cope with the case where the number of revolutions has increased. In addition, even with only rotation pulses, the position of the rotor cannot be accurately calculated because the position of the rotor can only be grasped in units of pitch angles due to the structure of the sensor gear. On the other hand, the motor control device according to the present invention includes a rotor position calculator that calculates the position of the rotor using the estimated position estimated by the position estimator and the rotation pulse. Therefore, when calculating the rotor position using the rotation pulse, the rotor position calculator can effectively use the estimated position of the position estimator to perform the calculation. That is, when the rotation of the motor is started, the rotor position calculator performs phase difference calculation processing for calculating the phase difference between the rotor position estimated by the position estimator and the rotor position based on the rotation pulse. conduct. This makes it possible to grasp where the reference position of the rotor exists within the range of the pitch angle. Therefore, in the rotor position calculation process, the rotor position calculator can accurately calculate the rotor position by calculating based on the phase difference in addition to the rotor position based on the rotation pulse. As described above, it is possible to accurately calculate the rotor position without providing an expensive device such as a resolver. Therefore, the position of the rotor can be calculated while reducing costs.

In a stage prior to the start-up process, the rotor position calculator calculates the initial value of the rotation pulse based on the correspondence relationship between the initial position of the rotor estimated by the position estimator and the step angle of the rotation pulse sensor. An initial value calculation process may be performed. In this case, the rotor position calculation section can calculate the rotor position after grasping the initial value of the pulse when the motor starts to rotate. Therefore, the rotor position can be easily calculated by the rotor position calculator.

In the phase difference calculation process, the rotor position calculation unit integrates the difference between the rotor position based on the rotation pulse and the estimated position each time the rotation pulse is counted after the motor is started, and calculates the average value of the integrated values. The phase difference may be calculated from In this case, the rotor position calculator can calculate the phase difference by comprehensively considering not only the difference when counting a single pulse, but also the difference for each count. Therefore, the phase difference can be calculated regardless of the manufacturing error for each tooth of the sensor gear.

In the rotor position calculation process, the rotor position calculation section may perform the rotor position calculation process of calculating the rotor position based also on the minute positional change amount of the rotor between pulses. The position of the rotor based on rotation pulses alone is constant because the position change between pulses is unknown (eg, see the dashed line graph in FIG. 8). On the other hand, in the rotor position calculation process, the rotor position calculation unit calculates the position change between pulses based on the minute position change amount of the rotor between the pulses, in addition to the rotor position based on the rotation pulses. can also be reflected to accurately calculate the rotor position.

In the rotor position calculation process, the rotor position calculator may calculate the minute position change amount by multiplying the angular velocity by the timer value measured from the time when the rotation pulse is counted. By using the angular velocity and the timer value in this way, it is possible to easily calculate the minute position change amount.

The rotor position calculation unit calculates the position of the rotor using the rotation pulse when the number of rotations is equal to or greater than a predetermined number of rotations, and switches so that the position estimated by the position estimation unit is used as the position of the rotor when the number of rotations is less than the predetermined number of rotations. may be performed. As a result, immediately after the start of rotation of the motor, while adopting the position estimated by the position estimator as the position of the rotor, by performing the above-described phase difference calculation processing and rotor position calculation processing, etc., the rotation pulse is used. Provisions can be made for the calculation of the rotor position. Then, after the number of revolutions has increased to some extent, the position of the rotor can be calculated more accurately than position estimation based on high-frequency superimposition by using the rotation pulse.

The rotor position calculator may synthesize the position of the rotor based on the rotation pulse and the estimated position by the position estimator. In this way, instead of discontinuously switching from position estimation based on harmonic wave superimposition to calculation using rotation pulses, the effects of calculation using harmonic superposition and calculation using rotation pulses gradually change. It is possible to perform continuous switching so as to continue.

ADVANTAGE OF THE INVENTION According to this invention, the motor control apparatus which can calculate the position of a rotor while reducing cost can be provided.

1 is a schematic configuration diagram showing a motor control device according to an embodiment of the present invention; FIG. It is a schematic diagram showing an example of a concrete composition of a motor. 3 is an enlarged view of a rotation pulse sensor; FIG. 4 is a flow chart showing control processing at the time of starting the motor by the motor control device according to the embodiment of the present invention; 7 is a graph showing changes in d-axis current and changes in estimation results over time. 4 is a graph showing the relationship between induced voltage and rotation pulse; 7 is a graph showing the relationship between the transition of the calculated rotor position based on the rotation pulse count value over time, the transition of the estimated position of the position estimator, and the rotation pulse. 7 is a graph showing time transition of calculation results by a rotor position calculation unit;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a schematic configuration diagram showing a motor control device 1 according to one embodiment of the present invention. The motor control device 1 according to the present embodiment is a device that can perform high-performance motor control without using a sensor such as a resolver for detecting an electrical angle position when starting the motor 2 . The motor control device 1 includes a motor 2, a current command section 3 (command section), a voltage command section 4 (command section), a conversion section 5, a detection section 6, a harmonic superposition section 7, and a rotation pulse sensor. 8 , a position estimator 9 , an angular velocity estimator 11 , a rotation pulse counter 12 , an angular velocity calculator 13 , a rotation pulse rotor position calculator 14 , and a rotor position calculator 15 .

The motor 2 is an electric motor driven by three-phase AC power. The motor 2 has a stator and a rotor with permanent magnets, and rotates the rotor. The motor 2 is provided with a rotation pulse sensor 8 for detecting rotation pulses of the motor 2 . An example of a specific configuration of the motor 2 is shown in FIG. As shown in FIG. 2, a U-phase, a V-phase, and a W-phase stator 22 and a rotor 20 are provided. Each stator 22 has an inductance portion 23 having a predetermined inductance and a resistance portion 24 having a predetermined resistance value. The rotor 20 has permanent magnets 21 . Each stator 22 rotates the rotor 20 by creating a rotating magnetic field around the rotor 20 when an alternating current is supplied. The motor 2 has R (resistance value), L (inductance), and _KE (induced voltage constant) as motor constants.

Here, for the motor 2, a rotating coordinate system synchronized with the rotor 20 and a fixed fixed coordinate system are set. In the rotating coordinate system, the main magnetic flux direction (N pole direction of the permanent magnet 21) is taken as the d-axis, and the q-axis orthogonal to the d-axis is taken. The fixed coordinate system has α and β axes that are orthogonal to each other. The angle formed by the d-axis of the rotating coordinate system with respect to the α-axis of the fixed coordinate system is defined as θ _re .

As shown in FIG. 1 , the current command section 3 and the voltage command section 4 function as command sections that transmit command signals to the motor 2 via the conversion section 5 . The current command unit 3 receives the rotation speed command, calculates a current command signal corresponding to the rotation speed command, and transmits the current command signal to the voltage command unit 4 . The current command unit 3 calculates a current command signal using speed PI control, for example. The current command unit 3 calculates a current command value (id command) on the _d -axis of the rotating coordinate system and a current command value (iq command) on the _q -axis of the rotating coordinate system.

Voltage command unit 4 receives a current command signal from current command unit 3 , calculates a voltage command signal corresponding to the current command, and transmits the voltage command signal to conversion unit 5 . The voltage command unit 4 calculates a voltage command signal using current PI control, for example. The voltage command unit 4 calculates a voltage command value on the d-axis and a current command value on the q-axis ( _iq command) based on a current command value (id command) on the _d -axis of the rotating coordinate system and a current command value (iq command) on the q-axis. The voltages on the q-axis are calculated and converted into voltage command values (Vu command, Vv command, Vw command) for each of the three-phase _U -phase, _V -phase, and _W -phase stators 22 .

The converter 5 is a device that applies an arbitrary alternating current to the stator 22 of the motor 2 . The converter 5 converts the voltage command value from the voltage commander 4 into a PWM signal. Further, the conversion unit 5 performs a switching operation based on the PWM signal to convert DC power into AC power, and applies a voltage corresponding to the voltage command calculated by the voltage command unit 4 to the motor 2 .

The detector 6 detects the current that flows through the motor 2 . The detector 6 also transmits the detected current to the voltage commander 4 and the position estimator 9 .

The harmonic superimposing unit 7 superimposes the harmonic on the command signal. The harmonic superimposing unit 7 superimposes harmonics on the current command signal from the current command unit 3 . The harmonic superimposing unit 7 superimposes a harmonic set to a predetermined frequency and a predetermined current value (amplitude). For example, as shown in FIG. 5, when the harmonic superimposing unit 7 superimposes a harmonic on a motor 2 commanded to supply a d-axis current of −50 A, the d-axis The current draws a waveform that oscillates with -50A as a reference. How to estimate the initial position of the rotor 20 by superimposing the harmonic superimposing section 7 will be described later.

The rotation pulse sensor 8 is a sensor that detects rotation pulses of the motor 2 . Specifically, as shown in FIG. 3 , the rotation pulse sensor 8 includes a sensor gear 16 and a detection section 17 . Sensor gear 16 has a plurality of teeth 18 provided at a predetermined pitch. Since the sensor gear 16 rotates in synchronization with the rotor 20 of the motor 2 , it rotates at the same rotational speed as the rotor 20 . The detection unit 17 is a device that detects rotation pulses generated as the sensor gear 16 rotates. The detection unit 17 is arranged at a position spaced apart from the sensor gear 16 on the outer peripheral side. When the sensor gear 16 rotates, each tooth 18 of the sensor gear 16 , which is a magnetic material, repeats the behavior of approaching and separating from the detection unit 17 . As a result, the state of the magnetic path in the detection section 17 changes, and the magnetic flux passing through the detection coil changes. Specifically, the magnetic flux increases as the tooth 18 approaches, and the magnetic flux decreases as the tooth 18 moves away. At this time, an induced electromotive force indicated by Faraday's law of electromagnetic induction is generated in the detection unit 17 . Rotation of the sensor gear 16 causes the magnetic flux penetrating the detection coil of the detection unit 17 to continuously increase and decrease, so that the electromotive force generated in the detection coil also increases and decreases while changing the polarity. The detector 17 detects such a power waveform as a rotation pulse. For example, of the edges 18a and 18b of the tooth 18, the waveform rises at the front edge 18a in the rotational direction, and the waveform falls at the rear edge 18b in the rotational direction (see the lower graph in FIG. 6).

A position estimator 9 estimates the position of the rotor 20 of the motor 2 . When the motor 2 is started, it is unknown at what electrical angle position (initial position) the rotor 20 is arranged in the direction of rotation. 20 initial positions are estimated. The position estimator 9 acquires the command signal for the motor 2 from the voltage commander 4, acquires the detection result from the detector 6, and estimates the initial position of the rotor 20 based on the information. If a harmonic is superimposed on the d-axis, a voltage corresponding to the harmonic is excited on the q-axis. position estimation. Although the position of the d-axis is unknown at the time of initial position estimation, the position estimator 9 estimates the initial position by a method described later. Further, the position estimator 9 estimates the position of the rotor 20 based on the high frequency superimposed signal even after the rotation of the motor 2 is started. The position estimator 9 outputs information on the estimated position to the rotor position calculator 15 .

The angular velocity estimator 11 estimates the angular velocity of the rotor 20 of the motor 2 . The angular velocity estimator 11 estimates angular velocity based on the estimated position estimated by the position estimator 9 . The angular velocity estimator 11 outputs information on the estimated angular velocity to the rotor position calculator 15 .

The rotation pulse counting section 12 counts rotation pulses detected by the rotation pulse sensor 8 . For example, when the rotation pulse sensor 8 outputs a signal of 32 pulses per rotation, the rotation pulse counting section 12 counts the rise and fall of the pulse, and counts 64 pulses per rotation. When the motor 2 is a two-pole motor, the relative position of the rotor 20 is grasped by the step angle of "360°/32=11.25° electrical angle" per count by the count of the rotation pulse counter 12. becomes possible. In the example of the rotation pulse shown in the lower part of FIG. 6, of the angles p1 and p2, the angle p1 corresponds to the step angle. The rotation pulse counting section 12 outputs the information on the count number of rotation pulses to the rotation pulse rotor position calculation section 14 .

The angular velocity calculator 13 calculates the angular velocity of the rotor 20 of the motor 2 . The angular velocity calculator 13 calculates angular velocity based on the rotation pulse detected by the rotation pulse sensor 8 . The angular velocity calculator 13 outputs information on the calculated angular velocity to the rotation pulse rotor position calculator 14 .

A rotation pulse rotor position calculation unit 14 acquires the count number of rotation pulses counted by the rotation pulse count unit 12 and calculates the position of the rotor 20 . The rotation pulse rotor position calculator 14 outputs the calculated rotor position and angular velocity to the rotor position calculator 15 .

The rotor position calculator 15 calculates the position of the rotor 20 using the estimated position estimated by the position estimator 9 and the rotation pulse. The rotor position calculator 15 acquires the estimated position from the position estimator 9 and the rotor position from the rotation pulse rotor position calculator 14, and performs calculations using these pieces of information to obtain the rotor 20 Calculate the actual rotational position of In addition, in order to distinguish the rotor position adopted by the rotor position calculation unit 15 from the rotor position calculated based on (only) the rotation pulse of the rotation pulse rotor position calculation unit 14, the term "actual rotor position" is used. sometimes referred to as Rotor position calculation unit 15 outputs the angular velocity to current command unit 3 .

In this embodiment, the rotor position calculator 15 calculates the actual rotor position of the rotor 20 based on the rotation pulse when the number of revolutions is equal to or greater than a predetermined number, and calculates the position estimated by the position estimator 9 when the number of revolutions is less than the predetermined number. is the actual rotor position. Therefore, from the initial state until the motor 2 is started, the rotor position calculator 15 uses the estimated position estimated by the position estimator 9 as the actual rotor position of the rotor 20 . Also immediately after the motor 2 is started, the rotor position calculator 15 adopts the estimated position estimated by the position estimator 9 as the actual rotor position of the rotor 20 while the number of revolutions is less than the predetermined number. On the other hand, when the number of revolutions is equal to or higher than the predetermined number, the rotor position calculation unit 15 switches to adopt the position of the rotor 20 calculated based on the rotation pulse by a method described later as the actual rotor position. Here, the rotor position calculator 15 uses the estimated position of the position estimator 9 to estimate the accuracy of the calculation of the actual rotor position based on the rotation pulse, in the stage prior to calculating the actual rotor position based on the rotation pulse. Perform pretreatment for lifting.

Referring to FIG. 4, the control processing performed by the motor control device 1 according to the present embodiment when the motor 2 is started and immediately after the start of rotation will be described. This control process is performed to calculate the actual rotor position of the rotor 20 without using a sensor such as a resolver at the stage of actually rotating the rotor 20 by starting the motor 2. processing. Further, the following description will be made with reference to FIGS. 5 to 8 as appropriate. FIG. 5 is a graph showing changes in d-axis current and changes in estimation results over time. The upper graph in FIG. 5 shows the d-axis current, and the lower graph shows the estimation results. In the example shown in FIG. 5, it is assumed that the actual electrical angle of the N pole of the rotor 20 is -270°. The graph of the estimated position is shown as a combination of straight lines, but in reality, there are places where fluctuations occur due to fluctuations in the estimation results and fluctuations in the detected values, but those fluctuations are omitted to facilitate understanding. is doing. FIG. 6 is a graph showing the relationship between the induced voltage and the rotation pulse. The graph on the upper side of FIG. 6 is a graph showing the induced voltage used for estimation by the position estimator 9 . The graph on the lower side of FIG. 6 is a graph showing rotation pulses. FIG. 7 is a graph showing the relationship between the transition of the calculation result of the rotor position based on the rotation pulse count value over time, the transition of the estimated position of the position estimator 9, and the rotation pulse. The graph on the upper side of FIG. 7 is a graph showing the transition of the calculation result of the rotor position based on the rotation pulse count value and the transition of the estimated position of the position estimator 9 over time. The graph on the lower side of FIG. 7 is a graph showing rotation pulses. FIG. 8 is a graph showing the time transition of the calculation result by the rotor position calculation unit 15. As shown in FIG.

As shown in FIG. 3, the motor control device 1 includes a position estimation process S10, an energization process S60, an initial value calculation process S70, a rotation start process S80, a phase difference calculation process S90, and a rotor position calculation process S110. and the switching process S120.

First, the motor control device 1 executes position estimation processing S10. The position estimating process S10 is a process in which the position estimating section 9 estimates the position of the rotor 20 with the harmonics superimposed thereon. In the position estimation process S10, the position estimation unit 9 executes an initial position assumption process S20, a determination process S30, and a confirmation process S40.

The motor control device 1 executes an initial position assumption process S20. The initial position assumption process S20 is a process in which the position estimator 9 assumes the initial position of the rotor 20 with the harmonics superimposed thereon. That is, since the position of the d-axis is unknown when the position estimation process S10 is started, it is necessary to assume an initial position. In the initial position assumption process S20, the current value commanded by the current command unit 3 is 0A. A harmonic superimposing unit 7 superimposes a harmonic on the current command signal. Therefore, as indicated by "S20" in the upper graph of FIG. 5, the d-axis current oscillates with 0A as a reference. In this state, the position estimator 9 assumes where the initial position of the rotor 20 is based on the command signal and the detection result of the detector 6 . At the stage of the initial position assumption processing S20, since the polarity of the permanent magnet of the rotor 20 is unknown, it is not possible to estimate the position of the north pole between 0 and 360 degrees. It can only be estimated as far as whether the pole or the south pole lies somewhere between 0 and 180 degrees. Therefore, when the position estimator 9 grasps any position between 0° and 180°, it assumes that the position is the initial position of the N pole. In the graph shown at the bottom of FIG. 5, the position of the north pole is assumed to be 120°.

Specifically, in the initial position assumption process S20, the harmonic superimposing unit 7 starts superimposing the harmonics with the number θ _re =0°. On the other hand, the position estimation unit 9 calculates extended induced voltages e _α and e _β in the αβ coordinate system, performs detection processing and bandpass filter processing, and calculates tan ⁻¹ (e _α /e _β ) as needed. Thus, the initial position of the north pole can be assumed.

The motor control device 1 executes determination processing S30. The determination process S30 is a process in which the position estimator 9 determines the magnetic poles of the rotor 20 based on the assumed initial position. In the discrimination process S30, the harmonic wave superimposing unit 7 applies positive/negative voltage pulses to the d-axis based on the assumed initial position. Then, the position estimator 9 determines the polarity of the permanent magnet 21 at the assumed initial position by observing the change in the d-axis current caused by the voltage pulse. Specifically, assuming that the 120° position is the d-axis, positive and negative voltage pulses are applied to the d-axis. A smaller d-axis current is the N pole, but in the lower graph of FIG. 4, the d-axis current fluctuates slightly toward the negative side. Therefore, the position estimator 9 determines that the magnetic pole present at the 120° position is the S pole.

The motor control device 1 executes a confirmation process S40. The determination process S40 is a process in which the position estimation unit 9 determines the initial position in the position estimation process S10 based on the determination result in the determination process S30. In the determination process S40, the position estimating unit 9 adds 180 degrees to the initial position of the N pole, which was assumed to be 120 degrees, and estimates that the initial position of the N pole is 300 degrees (the lower part of FIG. 5). S40 to S60 of the graph of ). Then, the harmonic superimposing unit 7 superimposes the harmonic with the estimated N pole, that is, the d-axis (φ=0) as a reference. As a result, the d-axis estimated position can be converged to determine the initial position. With the above, the position estimation process S10 is completed.

Next, the motor control device 1 executes an energization process S60. The energization process S60 is a process in which the current command unit 3 energizes the motor 2 with a d-axis current of a predetermined value based on the result of the position estimation process S10 in a state where the harmonics are superimposed by the harmonic superimposition unit 7. be. The predetermined value of the d-axis current is the current value when the motor 2 is normally stopped, that is, when the rotor 20 can be quickly rotated when the rotation speed command is received. In the upper graph of FIG. 5, the predetermined value of the d-axis current is set to -50 A, but it is not particularly limited.

The motor control device 1 executes an initial value calculation process S70. The initial value calculation process S70 is a step prior to the startup process S80. , the initial value of the rotation pulse is calculated. For example, the rotor position calculator 15 stores the quotient obtained by dividing the estimated initial position by the step angle as the initial value of the rotation pulse count. For example, if the estimated initial position is 68° and the increment angle is 11.25°, the rotation pulse The initial value of the count is stored as "6". As a result, when the motor 2 starts to rotate, it is possible to grasp how much the count of rotation pulses has progressed.

Next, the motor control device 1 executes a rotation starting process S80. The rotation starting process S80 is a process in which the current command unit 3 and the voltage command unit 4 start the rotation of the motor 2 in a state in which the initial position of the rotor 20 is estimated. Based on the estimated angular velocity and estimated position, the current command unit 3 generates the _id command and the _iq command, and the voltage command unit 4 generates the _Vu command, the _Vv command, and the _Vw command, Start the motor 2 to rotate.

Next, the motor control device 1 executes phase difference calculation processing S90. In the phase difference calculation process S90, the rotor position calculator 15 calculates the phase difference between the position of the rotor 20 estimated by the position estimator 9 and the position calculated by the rotation pulse rotor position calculator 14. is. In the phase difference calculation process S90, the rotor position calculator 15 integrates the difference between the position of the rotor 20 based on the rotation pulse and the estimated position each time the rotation pulse is counted after the motor 2 is started. Calculate the phase difference from the average value of the values.

Here, the position estimator 9 can continuously estimate the position. On the other hand, the rotation pulse is discontinuous, repeating rising and falling at predetermined intervals of electrical angle. Since it is not synchronized with the reference position (d-axis) of the rotor 20 that changes to , there is a phase difference between the estimated position and the position of the rotor based on the rotation pulse. occur.

Therefore, in order to calculate the phase difference, the rotor position calculator 15 obtains the difference between the position of the rotor 20 based on the rotation pulse and the estimated position. Specifically, the rotor position calculation unit 15 calculates the difference between the rotor position value based on the rotation pulse and the estimated position for each edge of the rotation pulse (that is, for each count value), thereby obtaining the count value. Calculate the phase difference at The rotor position calculation unit 15 can calculate the current count value by adding the count value counted at the start of rotation to the initial value acquired in the initial value calculation processing S70. Specifically, when the count value is 10 and the increment angle is 11.25°, the rotor position calculation unit 15 determines “count value: 10 × increment angle: 11.25° = based on rotation pulse. Rotor position: 112.5°” can obtain the rotor position based on the rotation pulse. If the position estimated by the position estimating unit 9 at this time is 120°, the rotor position calculating unit 15 outputs the following information: "estimated position: 120° - rotor position based on rotation pulse: 112.5° = phase difference: 7 .5°”, it can be calculated that the phase difference at “count value: 10” is 7.5°.

Here, as shown in the upper graph of FIG. 7, the rotor position calculated from the rotation pulse count value has a phase difference with respect to the position estimation estimated by the position estimator 9, Continue as is. Therefore, a phase difference occurs at each position on the horizontal axis, that is, at each count value. In the upper graph of FIG. 7, the phase difference is greatly deformed for easy understanding. If the time transition of the rotor position calculated from the count value of the rotation pulses is enlarged, it will have a shape like the graph indicated by the dashed line in FIG. That is, at the moment when a certain rotation pulse is counted, the rotor position value related to the calculation result rises sharply, and the rotor position value remains constant until the next rotation pulse is counted.

Here, due to the influence of manufacturing errors of the teeth 18 of the sensor gear 16, there are cases where the rotational pulse also slightly deviates for each count value. Therefore, after calculating the phase difference for one count value, the rotor position calculation unit 15 comprehensively judges the phase differences corresponding to a plurality of count values instead of fixing the value. confirm.

Specifically, the rotor position calculator 15 calculates the phase difference in each count value and integrates the phase differences. For example, when 6 is stored as the initial value of the count value, the rotor position calculator 15 calculates and integrates the phase differences for the count values 7 to 9 in the same manner as described above. Then, the rotor position calculator 15 newly adds the phase difference (7.5°) of the count value 10 to the integrated value. Similarly, the rotor position calculator 15 calculates and integrates the phase difference after the count value 11 as well. After accumulating the differences for the prescribed number of times, the rotor position calculation unit 15 obtains an average value from the accumulated values, and uses the average value as the phase difference between the rotation pulse and the position in the position estimation. Specifically, the rotor position calculation unit 15 calculates the phase difference 16 times. Integrated value: 118.4°/number of times: 16 times = average value: 7.4°", 7.4° can be obtained as the phase difference. Note that the number of times for obtaining the integrated value is not particularly limited. Further, the method of determining the phase difference from a plurality of phase differences is not limited to the method of calculating the average value from the integrated value, and may be calculated by other known statistical processing. Further, when ignoring manufacturing errors and the like, a value calculated from one count value may be determined as the phase difference.

The rotor position calculation process S110 is a process of calculating the position of the rotor 20 based on the minute rotation amount of the rotor 20 calculated from the position of the rotor 20 based on the rotation pulse, the phase difference, and the angular velocity. As shown in FIG. 8, the rotor position calculation unit 15 calculates the phase difference value indicated by the two-dot chain line and the minute rotation value indicated by the two-dot chain line with respect to the rotor position value based on the rotation pulse indicated by the broken line. is indicated by the solid line by adding the quantities . Compute the actual rotor position.

That is, as described above, the rotor position value based on the rotation pulse contains a phase difference with respect to the position of the estimated position. Therefore, in order to eliminate such a phase difference, the rotor position calculator 15 adds the value of the phase difference to the value of the rotor position based on the rotation pulse. Note that the value calculated in the phase difference calculation process S90 is used as the value of the phase difference.

Further, as described above, as shown in the dashed line graph in FIG. 8, the rotor position based on the rotation pulse increases stepwise based on the timing at which the rotation pulse is counted. On the other hand, the actual angle of the rotor 20 changes continuously with time. Therefore, it is necessary to add a minute position change amount (Δθ), which is a minute position change amount in a minute interval between pulses (see FIG. 8), of the rotor position based on the rotation pulses. Specifically, the rotor position calculation unit 15 multiplies the timer value measured from the time point when the rotation pulse is counted by the angular velocity (ω) to obtain "minute position change amount Δθ = angular velocity ω × timer A minute rotation amount is acquired by performing a calculation of "value Δt". As shown in FIG. 8, in the graph of the rotor position indicated by the broken line, the value increases over a minute rise time after counting the rotation pulses. Therefore, the minute position change amount graph indicated by the two-dot chain line is returned to 0 at the rising time. As for the timer value Δt, the measurement starts after the rise time has elapsed.

In the switching process S120, the rotor position calculation unit 15 changes the position estimated by the position estimation unit 9 to the actual rotor position (see FIG. 8) calculated based on the rotation pulse as the actual rotor position output to the command unit. This is the switching process. For example, the rotor position calculation unit 15 outputs the estimated position acquired from the position estimation unit 9 to the command units 3 and 4 to control the rotation of the motor 2 when the rotation speed is less than a predetermined number (for example, 100 rpm). On the other hand, the rotor position calculation unit 15 outputs the actual rotor position calculated in the rotor position calculation processing S110, that is, the actual rotor position calculated based on the rotation pulse when the predetermined number of revolutions is reached. Switching is performed so as to output to 3 and 4. When the switching is completed, there is no need to use the position estimated by the position estimator 9, so the rotor position calculator 15 stops the harmonic superimposing unit 7 from superimposing the harmonics.

In the case of deceleration of the motor 2, harmonic superposition is started from a predetermined number of revolutions (for example, 400 rpm), and the position estimator 9 obtains the estimated position. It may be switched to use the position estimated by the position estimating section 9 from the calculation result of the actual rotor position using the rotation pulse.

Next, functions and effects of the motor control device 1 according to the embodiment of the present invention will be described.

The motor control device 1 includes a position estimator 9 that estimates the position of the rotor by superimposing harmonics, and a rotation pulse sensor 8 that detects rotation pulses of the motor 2 . Here, if only the position estimated by the position estimating section 9 is used, it is not possible to cope with the case where the number of revolutions increases. Moreover, even with only the rotation pulse, the position of the rotor 20 cannot be accurately calculated because the position of the rotor 20 can be grasped only in units of pitch angles due to the structure of the sensor gear 16 . In contrast, the motor control device 1 according to the present embodiment includes a rotor position calculator 15 that calculates the position of the rotor 20 using the estimated position estimated by the position estimator 9 and rotation pulses. Therefore, when calculating the rotor position using the rotation pulse, the rotor position calculator 15 can effectively use the estimated position of the position estimator 9 to perform the calculation. That is, when the rotation of the motor 2 is started, the rotor position calculator 15 performs phase difference calculation processing to calculate the phase difference between the position of the rotor 20 estimated by the position estimator 9 and the position based on the rotation pulse. S90 is performed. This makes it possible to grasp where the reference position (d-axis) of the rotor 20 exists within the range of the pitch angle. Therefore, in the rotor position calculation process S110, the rotor position calculator 15 accurately calculates the actual rotor position by calculating based on the phase difference in addition to the position of the rotor 20 based on the rotation pulse. be able to. As described above, it is possible to accurately calculate the actual rotor position without providing an expensive device such as a resolver. Therefore, the position of the rotor can be calculated while reducing costs.

At the stage prior to the starting process S80, the rotor position calculation unit 15 calculates the initial position of the rotation pulse based on the correspondence relationship between the initial position of the rotor estimated by the position estimation unit 9 and the pitch angle of the rotation pulse sensor 8. An initial value calculation process S70 for calculating a value may be performed. In this case, the rotor position calculation unit 15 can calculate the actual rotor position after ascertaining the initial value of the pulse when the motor 2 starts rotating. Therefore, the rotor position can be easily calculated by the rotor position calculator 15 .

In the phase difference calculation process S90, the rotor position calculation unit 15 integrates the difference between the rotor position based on the rotation pulse and the estimated position each time the rotation pulse is counted after the motor is started, and calculates the integrated value. A phase difference may be calculated from the average value. In this case, the rotor position calculator 15 can calculate the phase difference by comprehensively considering not only the difference when counting a single pulse, but also the difference for each count. Therefore, the phase difference can be calculated regardless of the manufacturing error of each tooth 18 of the sensor gear 16 .

Also, the position of the rotor 20 based only on rotation pulses is constant because the position change between pulses is unknown (see, for example, the dashed line graph in FIG. 8). On the other hand, in the rotor position calculation processing S110, the rotor position calculation unit 15 calculates the position of the rotor 20 based on the rotation pulse, and also calculates the position of the rotor 20 based on the minute position change amount between the pulses. It is possible to accurately calculate the actual rotor position by also reflecting the change in the position of .

In the rotor position calculation process S110, the rotor position calculation unit 15 may calculate the minute position change amount by multiplying the angular velocity by the timer value measured from the time when the rotation pulse is counted. By using the angular velocity and the timer value in this way, it is possible to easily calculate the minute position change amount.

The rotor position calculation unit 15 calculates the position of the rotor using the rotation pulse when the number of rotations is equal to or greater than a predetermined number of rotations, and uses the position estimated by the position estimation unit as the position of the rotor when the number of rotations is less than the predetermined number of rotations. You can switch. As a result, immediately after the start of rotation of the motor 2, the position estimated by the position estimator 9 is used as the position of the rotor, and the above-described phase difference calculation processing S90 and rotor position calculation processing S110 are performed. Provision can be made for calculating rotor position using pulses. Then, after the number of revolutions has increased to some extent, the position of the rotor 20 can be calculated more accurately than position estimation based on high-frequency superimposition by using the rotation pulse.

The invention is not limited to the embodiments described above.

For example, in the position estimation process S10, the processes S20 to S40 are adopted, but the method is not limited to these methods. That is, other known initial position estimation methods may be applied to the position estimation process S10.

ｅ_α：α軸拡張誘起電圧
ｅ_β：β軸拡張誘起電圧
ｅ_ｄ：ｄ軸拡張誘起電圧
ｅ_ｑ：ｑ軸拡張誘起電圧
ｅ^Ｊθｒｅ：回転行列
θｒｅ：固定座標系のα軸に対して回転座標系のｄ軸がなす角度
ｅ_ｄωｒｅ：ｄ軸上の速度起因の拡張誘起電圧
ｅ_ｑωｒｅ：ｑ軸上の速度起因の拡張誘起電圧
ｅ_ｄｈ：ｄ軸上の高調波重畳起因の拡張誘起電圧
ｅ_ｑｈ：ｑ軸上の高調波重畳起因の拡張誘起電圧
ωre：電気角速度
Ｌ_ｄ：ｄ軸インダクタンス
Ｌ_ｑ：ｑ軸インダクタンス
Ｒ：抵抗
ｉ_ｄ：回転座標系のｄ軸上の電流
ｉ_ｑ：回転座標系のｑ軸上の電流
ドット付きのｉ_ｑ：回転座標系のｑ軸上の電流時間微分値
Ｋ_ｅ：誘起電圧定数

Further, the rotor position calculation unit 15 described above performs the switching process S120. However, instead of such switching, the rotor position calculator 15 may synthesize the rotor position based on the rotation pulse and the estimated position by the position estimator. For example, from the actual rotor position and motor parameters (R, Lq, Ld, etc.) calculated using the rotation pulse, eα and eβ in the fixed coordinate system are calculated, combined with the harmonic signal, and the harmonic A method of determining the signal amplitude for superimposition may be employed. For example, the extended induced voltage in the fixed coordinate system (α-β) is given by the following equations (1), (2), and (3). e _ωre is a speed-induced extended induced voltage, which is a term corresponding to the actual rotor position, and is expressed as in Equation (4). eh is an extended induced voltage caused by superimposition of harmonics, and is expressed as in Equation (5). θre, which indicates the position of the rotor, is given by Equation (6).

e _α : α-axis extended induced voltage e _β : β-axis extended induced voltage ed : _d -axis extended induced voltage e _q : q-axis extended induced voltage e ^Jθre : rotation matrix θre : coordinate rotated with respect to the α-axis of the fixed coordinate system Angle formed by the d-axis of the system e _dωre : extended induced voltage caused by velocity on the d-axis e _qωre : extended induced voltage caused by velocity on the q-axis e _dh : extended induced voltage caused by superimposition of harmonics on the d-axis e _qh : Extended induced voltage ωre caused by harmonic superimposition on the q-axis: Electrical angular velocity L _d : d-axis inductance L _q : q-axis inductance R: Resistance id : Current i q on the _d -axis of the rotating coordinate system i _q : Rotating coordinate system i _q with a current dot on the q-axis of : current-time differential value on the q-axis of the rotating coordinate system K _e : induced voltage constant

As the number of revolutions increases, the amplitudes of eα and eβ resulting from calculations based on rotation pulses become dominant over the amplitudes of eα and eβ due to high-frequency superimposition. In this way, instead of discontinuously switching from position estimation by high-frequency superimposition to calculation using rotation pulses, the influence of the calculation on harmonic superimposition and the calculation using rotation pulses gradually changes. It is possible to perform continuous switching as follows.

Also, the overall configuration of the motor control device 1 shown in FIG. 1 and the configuration of the motor 2 shown in FIG. 2 are merely examples, and can be changed as appropriate without departing from the scope of the present invention.

DESCRIPTION OF SYMBOLS 1... Motor control apparatus 2... Motor 3... Current command part (command part) 4... Voltage command part (command part) 7... Harmonic superposition part 8... Rotational pulse sensor 9... Position estimation part 15 ... Rotor position calculator, 20 ... Rotor, 21 ... Permanent magnet, 22 ... Stator.

Claims (7)

a motor comprising a stator and a rotor having permanent magnets;
a command unit that transmits a command signal to the motor;
a harmonic superimposing unit that superimposes a harmonic on the command signal;
a position estimating unit that estimates the position of the rotor by superimposing harmonics;
a rotation pulse sensor for detecting rotation pulses of the motor;
a rotor position calculator that calculates the position of the rotor using the estimated position estimated by the position estimator and the rotation pulse;
The command unit performs a starting process for starting rotation of the motor in a state in which the initial position of the rotor is estimated,
The rotor position calculation unit includes:
performing a phase difference calculation process for calculating a phase difference between the position of the rotor estimated by the position estimation unit and the position of the rotor based on the rotation pulse;
A motor control device that performs rotor position calculation processing for calculating the position of the rotor based on the position of the rotor based on the rotation pulse and the phase difference. In a stage prior to the start-up process, the rotor position calculation unit calculates the rotation pulse based on the correspondence relationship between the initial position of the rotor estimated by the position estimation unit and the step angle of the rotation pulse sensor. 2. The motor control device according to claim 1, which performs initial value calculation processing for calculating an initial value of . In the phase difference calculation process, the rotor position calculation unit integrates the difference between the rotor position based on the rotation pulse and the estimated position each time the rotation pulse is counted after the motor is started, 3. The motor control device according to claim 1, wherein said phase difference is calculated from an average value of integrated values. 3. In the rotor position calculation processing, the rotor position calculation section performs rotor position calculation processing for calculating the position of the rotor based on a minute positional change amount of the rotor between pulses. 4. The motor control device according to any one of 1 to 3. 3. In the rotor position calculation process, the rotor position calculation unit calculates the minute position change amount by multiplying a timer value measured from the time when the rotation pulse is counted by an angular velocity. 5. The motor control device according to 4. The rotor position calculation unit calculates the position of the rotor using a rotation pulse when the number of rotations is equal to or greater than a predetermined number of rotations, and calculates the position of the rotor by calculating the position estimated by the position estimation unit when the number of rotations is less than the predetermined number of rotations. 6. The motor control device according to any one of claims 1 to 5, wherein switching is performed such that The motor control device according to any one of claims 1 to 5, wherein the rotor position calculator synthesizes the position of the rotor based on the rotation pulse and the estimated position by the position estimator. .

Images