JP7321375B2 - motor controller - Google Patents

Description

The present disclosure relates to a motor control device that identifies motor parameters and inverter parameters.

Permanent magnet synchronous motors use vector control to improve control performance and reduce power loss during operation. A motor voltage equation representing the relationship between the dq-axis current and the dq-axis voltage in vector control is constructed using the armature winding resistance, inductance, and induced voltage constant of the permanent magnet synchronous motor as parameters.

In recent years, many sensorless methods have been adopted for estimating the position and rotational angular velocity of a rotor using dq-axis currents and dq-axis voltages without using position sensors. In the sensorless method, the rotor position and rotational angular velocity of the motor are estimated based on the motor voltage equation. Errors are a significant cause of sluggishness and step-out in the operation of permanent magnet synchronous motors.

Since the dead time reduces the output voltage, a voltage lower than the voltage command value by the dead time is applied to the motor, and a difference occurs between the voltage command value and the actual voltage applied to the motor. A voltage that decreases due to dead time is referred to herein as a dead time decreasing voltage. This dead time reduction voltage is added in advance to the voltage command value according to the direction (sign) of each phase motor current for compensation. Here, this compensation voltage is called dead time compensation voltage.

Dead time reduction voltage is a practical parameter. The dead time compensation voltage, on the other hand, is a parameter used by the controller.

Since the dead time varies depending on the individual characteristics of hardware such as switching elements and control circuit components, the dead time reduction voltage also varies. Therefore, even if the dead time compensation voltage is set to a specific value and compensation is performed, it is not possible to compensate for variations in the dead time reduction voltage. That is, an error occurs between the dead time compensation voltage, which is a parameter used by the controller, and the dead time reduction voltage, which is an actual parameter. As a result, a difference occurs between the voltage command value and the actual motor applied voltage. Here, this difference is called an output voltage error.

The output voltage error is non-linear with respect to the direction of the output current when viewed from a control device that controls the motor. Therefore, the output voltage error distorts the output current and degrades the control characteristics, resulting in torque ripple and increased motor noise and vibration. Also, power loss increases.

In order to prevent this as much as possible, conventionally, for dead time, for example, it was necessary to use expensive parts such as photocouplers as hardware, or to manually adjust power converters individually. rice field.

As one of the methods for solving this problem, a technique has been proposed in which current and voltage are input and the on-delay compensation value is automatically adjusted (see, for example, Patent Document 1).

The motor control device described in Patent Literature 1 measures a voltage command when the motor is stopped. The motor controller determines an on-delay compensation value by controlling the switching element so that constant DC currents flow at at least two switching frequencies. The two switching frequencies are hereinafter referred to as the first switching frequency and the second switching frequency. The first switching frequency is a switching frequency that is highly affected by dead time. The second switching frequency is a switching frequency less affected by dead time. Then, an on-delay compensation value is obtained based on the difference between the voltage command at the first switching frequency and the voltage command at the second switching frequency. According to this method, the on-delay compensation value can be obtained without being affected by errors in the motor resistance.

Also, the motor parameters used by the controller may have an error with respect to the actual parameters. As one of the methods for solving this problem, a technique for identifying motor parameters when the permanent magnet synchronous motor is stopped has been proposed (see, for example, Patent Document 2).

In the motor control device described in Patent Literature 2, after the rotor of the motor is stopped at a certain position, the superimposition identification unit identifies the armature winding resistance by the voltage equation expressed in the rotating coordinate system.

In the power converter control device described in Patent Document 3, the dead time is calculated based on the map storage means representing the voltage across the driving time of the switching element of the inverter, the passing current when driving, the temperature, and the single turn-off delay time characteristic. to correct. As a result, by eliminating the expected variation of the dead time, the dead time is sufficiently shortened and optimized while ensuring durability, and the output conversion function is improved.

Further, in Patent Document 3, the dead time is corrected based on the actual turn-off delay time measured by the delay time acquisition means and the turn-off delay time stored in the map storage means. As a result, even if there is individual variation in switching elements under the same conditions, the individual variation can be eliminated from the dead time, and the dead time can be further optimized.

Japanese Patent No. 3329831 JP 2018-186640 A JP 2010-142074 A

The technique described in Patent Document 1 does not identify motor parameters. Therefore, an error may occur in the armature winding resistance, which is one of the motor parameters. In the sensorless method, in which rotor position and rotational angular velocity are estimated based on motor voltage equations, errors in armature winding resistance cause a decrease in responsiveness and loss of synchronism in the operation of permanent magnet synchronous motors.

Further, in the identification, it is necessary to select a high frequency and a low frequency as the first and second switching frequencies in the technique described in Patent Document 1 above. Therefore, there are problems such as switching loss, heat generation, and noise associated therewith. Specifically, at high frequencies, there is a problem that switching loss increases and heat generation increases. Also, at lower frequencies the noise problem becomes more pronounced. In addition, since processing with two switching frequencies is required, there is also the problem that processing takes longer time than processing with one switching frequency.

Further, in the identification, in the technique described in the above-mentioned Patent Document 1, in order to prevent the AC motor from rotating, there is a ) is applied with a direct current. Furthermore, the on-delay compensation value is calculated using the current control output acquired at that time. This method has the problem that the stop position of the rotor of the AC motor, at which the on-delay compensation value can be calculated, is limited to the position where the DC current flows between the two phases, and cannot be an arbitrary position. In the technique of Patent Document 1, when the rotor of the AC motor must be stopped at a position other than the position where the DC current flows between the two phases, that is, the position where the DC current flows between the three phases, the on-delay compensation value is set to cannot be calculated.

On the other hand, the technique described in Patent Document 2 does not identify the dead time reduction voltage, which is one of the inverter parameters. Therefore, the method of identifying the armature winding resistance based on the motor voltage equation cannot identify it with high accuracy. As a result, in the sensorless method in which the rotor position and rotational angular velocity are estimated based on the motor voltage equation, the output voltage error and the armature winding resistance error affect the operational responsiveness of the permanent magnet synchronous motor. decrease and step-out are caused.

In the technique described in Patent Document 3, since map storage means is used to represent the voltage across the driving time of the switching element of the inverter, the passing current when driving, the temperature, and the single turn-off delay time characteristic, this characteristic is stored in advance. should be measured. The trouble of measurement is a burden on the user.

Moreover, in the technique described in Patent Document 3, a measuring device or a measuring circuit is used as the delay time acquisition means, so the product cost and product dimensions increase accordingly.

The present disclosure has been made to solve such problems, and enables identification of both motor parameters and inverter parameters, and aims to avoid a decrease in responsiveness and loss of synchronism of the motor. The aim is to obtain a control device.

A motor control device according to the present disclosure includes a current detection unit that detects a motor current flowing through a three-phase armature winding of a motor, a target rotational angular velocity input from the outside, and the motor current detected by the current detection unit. and a vector controller for controlling the actual rotational angular velocity of the motor by generating a voltage command value based on a power converter for supplying power to the motor according to the voltage command value generated by the vector controller. and an identifier that identifies motor parameters of the motor and inverter parameters of the power converter, wherein the identifier identifies the motor current of at least two of the three phases as a first value. stopping the rotor of the motor at a first position by controlling respectively to a second value and, with the rotor stopped at the first position, the voltage generated by the vector controller; A command value and the motor current detected by the current detection unit are obtained, and the motor parameter and the inverter parameter are obtained based on the voltage command value and the motor current. At the time of the second sampling, the rotor of the motor is moved to the first position by controlling the motor current of at least two of the three phases to the first value and the second value, respectively. With the rotor stopped at the first position, the voltage command value generated by the vector controller and the motor current detected by the current detection unit are obtained. and the first motor current, and at the second sampling, the motor current of at least two of the three phases is set to a fourth value different from the first value and the second value and a fifth value to stop the rotor of the motor at the second position, and with the rotor stopped at the second position, the voltage command value and the motor a second voltage command value and a second motor current, respectively, and the motor parameter based on the first and second voltage command values and the first and second motor currents. and the inverter parameter, the rotor of the motor by controlling the motor current of at least two phases out of the three phases to the first value and the second value, respectively is stopped at the first position for identification, and the motor current of at least two of the three phases is set to the first value and the second value, respectively. controlling to stop the rotor of the motor at the first position, and then, stepwise, to reduce the motor current of at least two of the three phases to the first value and the second value; A second identification method is a method of stopping the rotor of the motor at the second position by controlling the fourth value and the fifth value different from the At this time, the identifier presets an identification method for each position of the rotor from among the first identification method and the second identification method, and the identification method is set according to the position at which the rotor is stopped. Identification is performed by changing the identification method .

According to the motor control device according to the present disclosure, it is possible to identify both the motor parameter and the inverter parameter, thereby avoiding a decrease in responsiveness and loss of synchronism of the motor.

1 is a configuration diagram showing the configuration of a motor control device according to Embodiment 1; FIG. FIG. 5 is a diagram showing the relationship between three-phase dead time reduction voltage and motor current; 5 is a flowchart of automatic measurement processing of motor parameters and inverter parameters in the identifier of the motor control device according to Embodiment 1; FIG. 4 is a diagram showing a relationship between a d-axis phase and three-phase motor currents; The magnitude (absolute value) of the dead time reduction voltage is the same at each phase motor current value Iu_1 (>0), Iv_1 (>0), and Iw_1 (<0) when the rotor 10 is fixed. It is a figure which shows. If the "relationship between the motor currents (U-phase, V-phase, W-phase) and the dead time reduction voltage" is the same for all three phases in the saturation region, the phase motor currents Iu_1, Iv_1 and Iw_1 for positioning the rotor 10 are is in the saturation region. FIG. 5 is a table showing positive and negative motor currents and three-phase dead time reduction voltage values at (1), (3), (5), (7), (9) and (11) in FIG. 4; . Calculation formulas for calculating the d-axis dead time reduction voltage and the q-axis dead time reduction voltage in (1), (3), (5), (7), (9) and (11) of FIG. 4 are shown. It is a table. In (1), (3), (5), (7), (9) and (11) of FIG. It is a table|surface which shows the calculation formula for identifying. Calculation formulas for identifying the d-axis dead time reduction voltage ΔVd_1 and the q-axis dead time reduction voltage ΔVq_1 in (1), (3), (5), (7), (9) and (11) of FIG. is a table showing FIG. 5 is a table showing calculation formulas for identifying U-phase dead time reduction voltage ΔVu_1 in (1), (3), (5), (7), (9) and (11) of FIG. 4; FIG. 5 is a table showing calculation formulas for identifying V-phase dead time reduction voltage ΔVv_1 in (1), (3), (5), (7), (9) and (11) of FIG. 4; FIG. 5 is a table showing calculation formulas for identifying W-phase dead time reduction voltage ΔVw_1 in (1), (3), (5), (7), (9) and (11) of FIG. 4; FIG. 10 is a flowchart of automatic measurement processing of motor parameters and inverter parameters in the identifier of the motor control device according to Embodiment 2; FIG. 10 is a diagram showing a case where "relationship between motor currents (U-phase, V-phase, W-phase) and dead time reduction voltage" does not completely match for all three phases over the entire region; FIG. 10 is a diagram showing a case where the "relationship between motor currents (U-phase, V-phase, W-phase) and dead time reduction voltage" matches all three phases over the entire region; The d-axis phase θ_1 during the first positioning and the d-axis phase θ_2 during the second positioning are shown in (2), (4), (6), (8), (10) and (12) in FIG. ) and the values of the three-phase dead time reduction voltage at that time. When the d-axis phase θ_1 in the first positioning is (2), (4), (6), (8), (10) and (12) in FIG. 3 is a table showing values of three-phase dead time reduction voltages at various times; When the d-axis phase θ_2 in the second positioning is (2), (4), (6), (8), (10) and (12) in FIG. 3 is a table showing values of three-phase dead time reduction voltages at various times; 1 is a cross-sectional view showing an example of a configuration of a permanent magnet synchronous motor controlled by a motor control device according to Embodiment 1; FIG. 3 is a diagram showing an example of a configuration of a power converter in the motor control device according to Embodiment 1; FIG.

Hereinafter, embodiments of a motor control device and a motor control method according to the present disclosure will be described with reference to the drawings. The present disclosure is not limited to the following embodiments, and various modifications can be made without departing from the gist of the present disclosure. In addition, the present disclosure includes all possible combinations of the configurations shown in the following embodiments. Also, in each figure, the same reference numerals denote the same or corresponding parts, which are common throughout the specification. In each drawing, the relative dimensional relationship, shape, etc. of each component may differ from the actual one.

Embodiment 1.
FIG. 1 is a configuration diagram showing the configuration of a motor control device 100 according to Embodiment 1. As shown in FIG. As shown in FIG. 1 , the motor control device 100 controls a permanent magnet synchronous motor 101 .

FIG. 20 is a cross-sectional view showing an example of the configuration of permanent magnet synchronous motor 101 controlled by motor control device 100 according to the first embodiment. As shown in FIG. 20, the permanent magnet synchronous motor 101 is, for example, a three-phase AC motor having a stator 1 and a rotor 10. As shown in FIG. The stator 1 has a cylindrical stator core 2, one or more teeth 3 provided on the inner peripheral surface of the stator core 2, and an armature winding 4 wound around the teeth 3. there is Each tooth 3 is arranged at intervals in the circumferential direction. The armature winding 4 constitutes three-phase coils of a U-phase, a V-phase and a W-phase of the permanent magnet synchronous motor 101 . The rotor 10 is rotatably arranged inside the stator 1 . The rotor 10 has a rotor core 11 and one or more permanent magnets 12 embedded in the rotor core 11 . The permanent magnets 12 are arranged so that N poles and S poles alternate in the circumferential direction. The notations “N” and “S” shown in FIG. 20 indicate magnetic poles generated on the outer peripheral surface of rotor 10 . Note that FIG. 20 merely shows an example of the permanent magnet synchronous motor 101, and is not limited to this. That is, the number of permanent magnets 12, the number of teeth 3, and the like are appropriately set to arbitrary values. Also, the shapes of the stator 1, rotor 10, teeth 3, etc., and the arrangement of the armature winding 4 and the permanent magnets 12 can be arbitrarily changed.

As shown in FIG. 1, the motor control device 100 includes a current sensor 102, a three-phase/dq converter 104, an estimator 105, a speed controller 106, a current controller 107, and a dq/three-phase converter. 108 , a power converter 109 and an identifier 110 .

Of these, the three-phase/dq converter 104, the estimator 105, the speed controller 106, the current controller 107, and the dq/three-phase converter 108 control the actual rotational angular velocity and output of the permanent magnet synchronous motor 101. It constitutes a vector controller 120 that controls torque. Also, the vector controller 120 and the identifier 110 constitute a controller 130 .

The hardware configurations of the vector controller 120 and the identifier 110 will be briefly described. Each function of the vector controller 120 and each function of the identifier 110 are implemented by a processing circuit. The processing circuitry may be dedicated hardware or a processor executing a program stored in memory. If the processing circuitry is dedicated hardware, the processing circuitry may be implemented, for example, by a single circuit, multiple circuits, a programmed processor, or a combination thereof. When the processing circuit is a processor, the processing circuit is implemented by software, firmware, or a combination thereof. Software and firmware are written as programs and stored in memory. The processing circuit implements the function of each part by reading and executing the program stored in the memory.

Each part of the motor control device 100 will be described below.

The current sensor 102 constitutes a current detection section. Current sensor 102 includes three current sensors 102a, 102b and 102c. Current sensors 102a, 102b, and 102c detect armature currents flowing through the three-phase armature windings 4 (see FIGS. 20 and 21) of the permanent magnet synchronous motor 101. FIG. Specifically, the current sensor 102 a detects the armature current Iu flowing through the U-phase armature winding 4 of the permanent magnet synchronous motor 101 . The current sensor 102 b detects an armature current Iv flowing through the V-phase armature winding 4 of the permanent magnet synchronous motor 101 . The current sensor 102 c detects an armature current Iw flowing through the W-phase armature winding 4 of the permanent magnet synchronous motor 101 . Armature currents Iu, Iv and Iw are hereinafter collectively referred to as three-phase motor currents Iu, Iv and Iw. When calling each phase separately, they are called motor current Iu, motor current Iv, and motor current Iw.

Three-phase/dq converter 104 converts three-phase motor currents Iu, Iv and Iw obtained by current sensors 102a, 102b and 102c into d-axis current Id and q-axis current Iq on dq coordinates.

A target rotation angular velocity ω ^* is input to the speed controller 106 from the outside. A speed controller 106 generates a d-axis current command value Id ^* and a q-axis current command value Iq ^* on the dq coordinates based on the difference between the target rotation angular velocity ω ^* and the actual rotation angular velocity ω. Here, in the dq coordinates, the magnetic field direction of the rotor 10 of the permanent magnet synchronous motor 101 is defined as the d-axis, that is, the magnetic field axis, and the direction orthogonal to the d-axis is defined as the q-axis. Note that when the actual rotational angular velocity ω can be obtained by measurement, that value may be used. Further, when the actual rotational angular velocity ω cannot be obtained, the estimated value ωest of the actual rotational angular velocity ω output from the estimator 105 is used as the actual rotational angular velocity ω. The estimated value ωest of the actual rotational angular velocity ω will be described later.

Based on the deviation between the d-axis current Id and the q-axis current Iq and the d-axis current command value Id ^* and the q-axis current command value Iq ^* , the current controller 107 adjusts the d-axis voltage command value Vd ^* and the q-axis voltage command value. Generate the value Vq ^* .

A dq/three-phase converter 108 converts the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* into three-phase voltage command values Vu ^* , Vv ^* and Vw ^* .

Power converter 109 supplies voltage command values Vu ^** after dead time compensation, which are obtained by adding three-phase dead time compensation voltages Vucomp, Vvcomp, and Vwcomp to three-phase voltage command values Vu ^* , Vv ^* , and ^Vw* . Vv ^** and Vw ^** are entered. Power converter 109 subtracts three-phase dead-time voltages ΔVu, ΔVv, and ΔVw from voltage command values Vu ^** , Vv ^** , and Vw ^** , and converts three-phase voltages Vu, Vv, and Vw to permanent magnet synchronous voltages. Output to motor 101 .

Note that the power converter 109 is composed of a switching circuit. FIG. 21 is a diagram showing an example of the configuration of power converter 109 in motor control device 100 according to the first embodiment. Power converter 109 has a plurality of switching elements 20 and 21, as shown in FIG. Output Vu, Vv and Vw. An upper-arm switching element 20 and a lower-arm switching element 21 are provided for each of the U-phase, V-phase, and W-phase. The switching element 20 of the upper arm and the switching element 21 of the lower arm are connected in series. A connection point 22 between the switching element 20 of the upper arm and the switching element 21 of the lower arm is connected to the armature winding 4 of each phase of the permanent magnet synchronous motor 101 . Semiconductor switching elements such as IGBTs are used for the switching elements 20 and 21, for example. A free wheel diode 25 is connected in anti-parallel to each of the switching elements 20 and 21 as shown in FIG.

As shown in FIG. 21, power converter 109 is connected to power supply 30 . A capacitor 31 is connected in parallel with the power converter 109 . Capacitor 31 suppresses current fluctuations caused by switching operations between power supply 30 and power converter 109 . The power supply 30 is composed of a DC power supply.

Power converter 109 receives a drive signal from the control application connected to gate 23 of each switching element 20 and 21 . The drive signal is a PWM pattern consisting of an ON signal and an OFF signal. Also, the control application is executed by, for example, a microcomputer. Power converter 109 converts the DC voltage applied by power supply 30 into three-phase voltages Vu, Vv and Vw by turning on and off switching elements 20 and 21 according to the drive signal. Three-phase voltages Vu, Vv and Vw are output to permanent magnet synchronous motor 101 . In this manner, the power converter 109 drives the permanent magnet synchronous motor 101, which is an object to be controlled. The microcomputer is equipped with a peripheral circuit section for driving the permanent magnet synchronous motor 101 . The power converter 109 converts the three-phase motor currents Iu, Iv, and Iw obtained by the three current sensors 102a, 102b, and 102c into digital signals using an AD conversion unit, which is one of those peripheral circuit units. to perform feedback control.

In Embodiment 1, a voltage source inverter is used for power converter 109 . A voltage source inverter is a device that switches a DC voltage supplied from a power supply 30 and converts it into an AC voltage. Note that the power converter 109 is not limited to a voltage source inverter, as long as it can output AC power for driving the permanent magnet synchronous motor 101. A circuit such as a matrix converter that converts to , or a multi-level converter in which outputs of a plurality of converters are connected in series or in parallel may be used.

The identifier 110 shown in FIG. 1 holds motor voltage information, which defines the relationship between the actual voltage of the permanent magnet synchronous motor 101, the motor current, and the armature winding resistance, which will be described later. The actual voltages are, for example, three-phase voltages Vu, Vv and Vw, or dq-axis voltages Vd and Vq. Here, the latter case will be described as an example. Therefore, the motor voltage information held by the identifier 110 defines the relationship between the dq-axis voltage, the dq-axis current, and the armature winding resistance of the permanent magnet synchronous motor 101 . The motor voltage information is prestored in memory, for example. The identifier 110 identifies the armature winding resistance of the armature winding 4 of the permanent magnet synchronous motor 101 and the dead time reduction voltage. When performing identification, the identifier 110 first controls the phase currents of at least two of the three phases to the first value and the second value, respectively, so that the permanent magnet synchronous motor 101 rotates. The child 10 is stopped at the preset first position. The identifier 110 obtains the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* generated by the current controller 107 with the rotor 10 stopped at its first position. The identifier 110 also obtains the d-axis current Id and the q-axis current Iq converted to dq coordinates by the three-phase/dq converter 104 . The d-axis current Id and the q-axis current Iq are three-phase/dq-converted three-phase motor currents Iu, Iv, and Iw detected by the current sensor 102 . The d-axis voltage Vd and the q-axis voltage Vq of the permanent magnet synchronous motor 101 are obtained by adding the d-axis dead time compensation voltage Vdcomp and the q-axis dead time compensation voltage Vqcomp to the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* . is added and the d-axis dead time reduction voltage ΔVd and the q-axis dead time reduction voltage ΔVq are subtracted. Therefore, the identifier 110 uses this to perform the following processing. That is, the identifier 110 uses the d-axis voltage command value Vd ^* , the q-axis voltage command value Vq ^* , the d-axis current Id and the q-axis current Iq, and the motor voltage information to determine the armature winding of the armature winding 4. A line resistance Ra, a d-axis dead time reduction voltage ΔVd, and a q-axis dead time reduction voltage ΔVq are obtained. Details of the operation of the identifier 110 will be described later. The d-axis dead time compensation voltage Vdcomp and the q-axis dead time compensation voltage Vqcomp on the dq coordinates are three-phase/dq-converted three-phase dead time compensation voltages Vucomp, Vvcomp, and Vwcomp.

Here, the motor voltage information is, for example, a motor voltage equation represented by equation (1) described later, or one of equations (2) to (25) described later derived by modifying the motor voltage equation. At least one. Furthermore, the motor voltage information may include at least one of values and calculation formulas shown in FIGS. Numerical data obtained by the motor voltage equation and the above calculation formula or numerical data obtained by measurement may be stored in memory in advance as a data table, and the data table may be used as the motor voltage information. . Further, values and calculation formulas shown in FIGS. 7 to 13 and FIGS. 17 to 19, which will be described later, may be stored in memory in advance as a data table, and the data table may be used as the motor voltage information.

Estimator 105 receives d-axis voltage command value Vd ^* and q-axis voltage command value Vq ^* from current controller 107 . Estimator 105 also receives d-axis current Id and q-axis current Iq from three-phase/dq converter 104 . The estimator 105 estimates the actual rotational angular velocity ω of the permanent magnet synchronous motor 101 based on the input d-axis voltage command value Vd ^* and q-axis voltage command value Vq ^* and the d-axis current Id and q-axis current Iq. Determine the value ωest and the position. An estimated value ωest of the actual rotational angular velocity ω is input to the speed controller 106 . The position is also converted into a position (electrical angle) (d-axis phase) and input to the dq/three-phase converter 108 and the three-phase/dq converter 104 .

In Embodiment 1, it is assumed that vector controller 120 performs so-called sensorless vector control. In the sensorless control, the current and voltage on the dq coordinates and the parameters of the permanent magnet synchronous motor 101 are used to determine the position of the rotor 10 and the actual rotational angular velocity ω to estimate In the field of sensorless vector control, the dq-axis is sometimes referred to as the γδ-axis, but in the following description of Embodiment 1, the dq-axis and the γδ-axis are used without distinction .

An outline of the operation of the vector controller 120 when the permanent magnet synchronous motor 101 is driven will be described. As described above, the vector controller 120 is composed of the 3-phase/dq converter 104, the estimator 105, the speed controller 106, the current controller 107, and the dq/3-phase converter 108. .

First, as shown in FIG. 1, the speed controller 106 receives the difference between the externally input target rotation angular velocity ω ^* and the estimated value ωest of the actual rotation angular velocity ω estimated by the estimator 105 . Based on the difference, speed controller 106 generates d-axis current command value Id ^* and q-axis current command value Iq ^* , respectively.

Three-phase motor currents Iu, Iv and Iw obtained by current sensors 102 a , 102 b and 102 c are input to three-phase/dq converter 104 . Three-phase/dq converter 104 converts three-phase motor currents Iu, Iv and Iw into d-axis current Id and q-axis current Iq. Current controller 107 receives the deviation between d-axis current Id and d-axis current command value Id ^* and the deviation between q-axis current Iq and q-axis current command value Iq ^* . Current controller 107 generates d-axis voltage command value Vd ^* and q-axis voltage command value Vq ^* based on those deviations.

The d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* are input to the dq/three-phase converter 108 . A dq/three-phase converter 108 converts the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* into three-phase voltage command values Vu ^* , Vv ^* and Vw ^* . Power converter 109 supplies voltage commands Vu ^** , ^{Vv *} after dead time compensation, which are obtained by adding three-phase dead time compensation voltages Vucomp, Vvcomp, and Vwcomp to three-phase voltage command values Vu ^* , Vv*, and Vw ^* . ^* and Vw ^** are entered. Power converter 109 generates three-phase voltages Vu, ^{Vv and Vw by subtracting three-phase dead time reduction voltages ΔVu, ΔVv and ΔVw from voltage commands Vu** ,} Vv** and Vw ^** , respectively. The generated three-phase voltages Vu, Vv and Vw are output to permanent magnet synchronous motor 101 . In this manner, the permanent magnet synchronous motor 101 is controlled such that the actual rotational angular velocity ω follows the target rotational angular velocity ω ^* .

Here, the dead time will be briefly explained. As described with reference to FIG. 21, the switching circuit is configured by connecting the upper arm switching element and the lower arm switching element in series. Therefore, when the switching elements of the upper and lower arms are turned on at the same time, the switching elements of the upper and lower arms are short-circuited and destroyed. In order to prevent this short circuit, a period is provided in which the switching elements of the upper and lower arms are both turned off when the switching elements are switched on and off. This period is called dead time td (tdu, tdv, tdw).

Since the output voltage decreases due to this dead time, a voltage lower than the voltage command value by the dead time is applied to the motor, causing a difference between the voltage command value and the actual voltage applied to the motor. The voltage that decreases due to this dead time is the dead time decreasing voltage, as described above.

The three-phase dead time reduction voltages ΔVu, ΔVv and ΔVw are shown in the following equation (100).
ΔVu=Vdc×tdu×fsw×sign(Iu)
ΔVv=Vdc×tdv×fsw×sign(Iv) (100)
ΔVw = Vdc x tdw x fsw x sign (Iw)
Here, sign(i)=1 (when i>0), 0 (when i=0), -1 (when i<0). In addition, Vdc is the inverter DC bus voltage, fsw is the carrier frequency, tdu, tdv, and tdw are the dead times of each phase. ing.

For example, regarding the U phase, "Vdc×tdu×fsw" in equation (100) is the magnitude of the dead time reduction voltage ΔVu. This decrement "Vdc.times.tdu.times.fsw" is subtracted from voltage command Vu ^** after dead time compensation according to the direction (sign) of the U-phase motor current. That is, if the motor current is positive, +Vdc×tdu×fsw is subtracted, and if the motor current is negative, −Vdc×tdu×fsw is subtracted from the voltage command Vu ^** after dead time compensation. As can be seen from equation (100), the magnitudes of the dead time reduction voltages ΔVu, ΔVv and ΔVw depend only on the inverter DC bus voltage, carrier frequency and dead time. For details of the dead time reduction voltage, refer to, for example, Non-Patent Document 1 (Non-Patent Document 1: Jun Kudo, Toshihiko Noguchi, Manabu Kawakami, Koichi Sano, "Mathematical model error of IPM motor control system and its compensation Law", Semiconductor Power Conversion Study Group, January 2008, SPC-08-25).

In the first embodiment, the voltages corresponding to the dead time reduction voltages ΔVu, ΔVv and ΔVw are set as the dead time compensation voltages Vucomp, Vvcomp and Vwcomp according to the direction (sign) of the motor current of each phase as voltage command values Vu ^* , Vv ^* and Vw ^* are pre-added to compensate. Note that the dead time reduction voltage is an actual parameter. The dead time compensation voltage, on the other hand, is a parameter used by the controller 130 .

Dead time compensation voltages Vucomp, Vvcomp and Vwcomp are shown in the following equation (101).
Vucomp = Vdc x tducomp x fsw x sign (Iu)
Vvcomp=Vdc*tdvcomp*fsw*sign(Iv) (101)
Vwcomp = Vdc x tdwcomp x fsw x sign(Iw)
Here, tducomp, tdvcomp, and tdwcomp are dead time compensation amounts for each phase. It is optimal to set the dead time td (tdu, tdv, tdw) as the dead time compensation amount tdcomp (tducomp, tdvcomp, tdwcomp). Also, sign(i)=1 (when i>0), 0 (when i=0), and -1 (when i<0).

For example, for the U phase, add +Vdc×tducomp×fsw if the motor current is positive, or -Vdc×tducomp×fsw if the motor current is negative, to the U phase voltage command value Vu ^* to reduce the voltage due to dead time. make up for In this way, dead time compensation is used to eliminate the effects of dead time. This dead time compensation is performed by adding a dead time compensation amount tdcomp (tducomp, tdvcomp, tdwcomp) considering the direction in which the motor current I (Iu, Iv, Iw) flows to the ON signal time Ton input from the control application. You may compensate for the dead time as follows.
T'on ← Ton + tdcomp x sign(I) (102)
For details of dead time compensation using equation (102), refer to Patent Document 1, for example.

However, the dead time td is the time ton from inputting an on signal to the switching element until it actually turns on, and the time toff from inputting an off signal to actually turning off the switching element. Let dead time tdead be the time during which the driving signals input from the application are turned off at the same time.
td = tdead + ton - toff
becomes. Since the times tdead, ton, and toff fluctuate or vary depending on individual characteristics of hardware such as switching elements and control circuit components, td also fluctuates or varies. Therefore, if a certain specified value is used as the dead time compensation voltage, an error may occur between the dead time compensation voltage and the dead time reduction voltage. That is, an error occurs between the parameters used by the controller 130 and the actual parameters. As a result, an output voltage error occurs between the voltage command value and the actual voltage applied to the motor.

As described above, the output voltage error is non-linear with respect to the direction of the output current and the like when viewed from the control device that controls the motor. Therefore, the output voltage error distorts the output current and degrades the control characteristics, resulting in torque ripple and increased motor noise and vibration. Also, power loss increases.

As one of the methods to solve this problem, the dead time reduction voltages ΔVu, ΔVv and ΔVw are identified, and the identification results are used as the dead time compensation voltages Vucomp, Vvcomp and Vwcomp to reduce the output voltage error to zero or within the allowable range. technology to

This technique will be described in detail. As described above, as can be seen from equation (100), the magnitude of the deadtime reduction voltage depends only on the inverter DC bus voltage, carrier frequency, and deadtime.

If the dead times tdu, tdv, tdw and the dead time compensation amounts tducomp, tdvcomp, tdwcomp do not match, a dead time error, which is the difference between them, occurs. The dead time error includes predictable variations and unpredictable variations. Predictable fluctuations include fluctuations predicted from the motor current of each phase, the voltage across the switching element when it is driven, and the temperature. Unpredictable variations include variations due to individual characteristics of hardware such as switching elements and control circuit components.

Thus, the dead-time error causes an output voltage error. In order to solve this problem, the identifier 110 stores in memory in advance the relationship between each phase motor current, the voltage across the switching element when driving, and the temperature and each phase dead time reduction voltage ΔVu, ΔVv, ΔVw. need to keep In that case, the identifier 110 reads out these relationships stored in the memory, and based on the actual motor current, the voltage across the driving switching element, and the temperature, the dead time reduction voltages ΔVu, ΔVv, ΔVw can be identified. In order to hold these relationships in memory in advance, it is necessary to measure these relationships in advance. However, the trouble of the measurement becomes a burden on the user. For a method of identifying the dead time by holding these relationships in memory, see Patent Document 3, for example.

Even if the variation of the dead time error predicted from the actual motor current, the voltage across the switching element driving, and the temperature is completely eliminated, the individual characteristics of the hardware such as the switching element and control circuit components , the dead time error varies. In order to prevent this as much as possible, conventionally, for example, regarding dead time, expensive parts such as photocouplers are used as hardware, or each phase motor current, switching element drive time It is necessary to measure the voltages across the terminals and the relationship between the temperature and the dead time reduction voltages ΔVu, ΔVv, and ΔVw of each phase and store them in memory in advance. Such measurements are required for the number of power converters, which imposes a considerable burden on the user.

Therefore, in Embodiment 1, the identifier 110 performs identification without previously measuring the dead time reduction voltages ΔVu, ΔVv, and ΔVw. The identification method is described below. Further, by performing the identification, it becomes possible to deal with the dead time error variation and variation. As a result, the burden on the user is eliminated.

In the above description, the error occurring between the dead time compensating voltage and the dead time reduction voltage is taken as an example of the factor that causes the output voltage error. Another factor is the error due to the ON voltage of the switching element. The error caused by the ON voltage and its compensation method are described in Non-Patent Document 1 and the like. If the on-voltage must be taken into consideration, the output voltage error can be made almost zero by compensating by the method described in Non-Patent Document 1 or the like. Therefore, the error due to the ON voltage is not taken into account here.

The operation of the identifier 110 will be described in detail with reference to FIGS. 3 to 13. FIG. FIG. 3 is a flowchart of automatic measurement processing of motor parameters and inverter parameters in the identifier 110 of the motor control device 100 according to the first embodiment. 4 to 13 will be described later.

The process of FIG. 3 is performed by the identifier 110 as an application for obtaining motor parameters of the permanent magnet synchronous motor 101 and inverter parameters of the power converter 109 . The processing in FIG. 3 can be performed when a command to start the permanent magnet synchronous motor 101 is given. Further, the motor control device 100 can drive the permanent magnet synchronous motor 101 using the motor parameters and the inverter parameters obtained in the processing of FIG. The motor parameter identified by the process of FIG. 3 is the armature winding resistance Ra of the armature winding 4, and the inverter parameters are the three-phase dead time decreasing voltages ΔVu_1, ΔVv_1 and ΔVw_1. Note that the inverter parameter is not limited to this, and may be, for example, the d-axis dead time reduction voltage ΔVd_1 and the q-axis dead time reduction voltage ΔVq_1.

When starting the series of processes shown in FIG. 3, the identifier 110 first rotates the rotor 10 to a preset first position by controlling the dq-axis currents in steps S100 and S101. Positioning operation is performed by moving and fixing. The positioning operation is performed in a state where the rotor 10 of the permanent magnet synchronous motor 101 is stationary, in order to acquire the voltage command value necessary for identifying the motor parameters and identifying the inverter parameters.

To describe the positioning operation in detail, first, in step S100, in order to position the rotor 10, the identifier 110 sends a current command to the dq axes of the rotating coordinate system of the permanent magnet synchronous motor 101 and the rotating coordinate Give the position (d-axis phase) θe relative to the reference position of the system. As the current command to be given, a preset DC value Id ^* _1 is set for the d-axis current command value, which is the field axis, and Iq ^* _1 (=0), which is zero, is set for the q-axis current command value. . Also, θ_1 is set as the d-axis phase. Then, during the execution period of the voltage command value acquisition process in step S102, the identifier 110 obtains the DC value Id ^* _1 and the q-axis current command value Iq ^* _1 (=0) so that the rotor 10 does not move. , and the d-axis phase θ_1. At this time, the DC value Id ^* _1 and the q-axis current command value Iq ^* _1 (=0) are set by the identifier 110, but the speed controller 106 may be set. Also, although the d-axis phase θ_1 is set by the identifier 110, it may be set by the estimator 105 as well.

FIG. 4 is a diagram showing the relationship between the d-axis phase θe and the three-phase motor currents Iu, Iv and Iw. In FIG. 4, the horizontal axis indicates the d-axis phase θe, and the vertical axis indicates the three-phase motor currents Iu, Iv and Iw. The position of the U-phase armature winding 4 is the reference position (θe=0) of the d-axis phase θe. In FIG. 4, the solid line indicates the U-phase motor current Iu, the dashed line indicates the V-phase motor current Iv, and the dotted line indicates the W-phase motor current Iw. As shown in FIG. 4, the three-phase motor currents Iu, Iv and Iw are sinusoidal waves that are out of phase with each other by 120°.

At this time, the d-axis current command value Id ^* _1 and the q-axis current command value Iq ^* _1 are set so that, for example, a DC current of Ia flows in the U phase, -Ia/2 in the V phase, and -Ia/2 in the W phase. Assume that (=0) is set. In this case, the rotor 10 is fixed at the position of θ_1=0, which is the reference position. Of course, the position where the rotor 10 is fixed is not particularly limited to this.

Here, in the phases within the ranges indicated by the six arrows (1), (3), (5), (7), (9) and (11) in FIG. and Iw, none of the motor currents becomes zero.

In step S100, the identifier 110 performs Set the d-axis phase θ_1.

In step S101, convergence of the d-axis current is completed, and positioning of the rotor 10 is completed.

Next, in step S102, the identifier 110 acquires the d-axis voltage command value Vd ^* _1 and the q-axis voltage command value Vq ^* _1 from the current controller 107, and obtains from the three-phase/dq converter 104 on the dq coordinates Obtain the d-axis current Id_1 and the q-axis current Iq_1. At this time, d-axis current command ^{value Id*_1 and q-axis current command value Iq* _1} may be obtained from speed controller 106 instead of d-axis current Id_1 and q-axis current Iq_1.

When sampling of d-axis voltage command value Vd ^* _1, q-axis voltage command value Vq ^* _1, and d-axis current Id_1 and q-axis current Iq_1 in step S102 is completed, the process of identifier 110 proceeds to step S103.

In step S103, the identifier 110 starts identifying the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1. Then, in step S104, the identifier 110 calculates the armature winding resistance Ra of the armature winding 4 and the dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1.

Here, regarding the method of identifying the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu_1, ΔVv_1, and ΔVw_1, the motor voltage equation and the "each phase motor current and dead time reduction voltage”.

Generally, the motor voltage equation of a permanent magnet synchronous motor is given by the following equation (1). In equation (1), Ra is the armature winding resistance of the armature winding 4, Ld is the d-axis inductance, Lq is the q-axis inductance, ωe is the actual rotational angular velocity (electrical angle), and Ψa is the magnetic flux chain by the permanent magnet 12. The intersection, p, represents the differential operator.

The d-axis voltage Vd and the q-axis voltage Vq on the dq coordinates on the left side of Equation (1) are the three-phase/dq-converted three-phase voltages Vu, Vv, and Vw.

These three-phase voltages Vu, Vv, and Vw are obtained by adding three-phase dead time compensation voltages Vucomp, Vvcomp, and Vwcomp to three-phase voltage command values Vu ^* , Vv ^* , and Vw ^* to obtain three-phase dead time reduction voltages ΔVu, ΔVv. and ΔVw.

The d-axis dead time compensation voltage Vdcomp and the q-axis dead time compensation voltage Vqcomp on the dq coordinates are three-phase/dq-converted three-phase dead time compensation voltages Vucomp, Vvcomp, and Vwcomp. Also, the d-axis dead time reduction voltage ΔVd and the q-axis dead time reduction voltage ΔVq on the dq coordinates are three-phase/dq-converted three-phase dead time reduction voltages ΔVu, ΔVv, and ΔVw.

Therefore, the d-axis voltage Vd and the q-axis voltage Vq are obtained by applying the d-axis dead time compensation voltage Vdcomp and the q-axis dead time compensation voltage Vqcomp to the d-axis voltage command value ^Vd* and the q-axis voltage command value Vq ^* on the dq coordinates. are added and the d-axis dead time reduction voltage ΔVd and the q-axis dead time reduction voltage ΔVq are subtracted.

Let Vd_1 and Vq_1 be the actual voltages applied to the motor on the dq coordinates at the first sampling, ΔVdcomp_1 and ΔVqcomp_1 be the dead time compensation voltages, and ΔVd_1 and ΔVq_1 be the dead time reduction voltages. Applying, the actual motor applied voltages Vd_1 and Vq_1 can be expressed as Equations (4) and (5), respectively.

Note that the actual motor-applied voltages Vd_1 and Vq_1 on the dq coordinates are the three-phase/dq-converted three-phase voltages Vu_1, Vv_1, and Vw_1 actually applied to the motor 101 during the first sampling. Also, the d-axis dead time compensation voltage Vdcomp_1 and the q-axis dead time compensation voltage Vqcomp_1 on the dq coordinates are three-phase/dq-converted three-phase dead time compensation voltages Vucomp_1, Vvcomp_1, and Vwcomp_1 at the time of the first sampling. be.

In the first embodiment, the d-axis current command value Id ^* and the q-axis current command value Iq ^* , which are DC current commands, are applied during the process of identifying the motor parameters and the inverter parameters, and the permanent magnet synchronous motor 101 Since the rotor 10 is stationary, the rotational angular velocity (electrical angle) ωe is ωe=0.

Therefore, the voltage equation on the d-axis of Equation (1) is represented by Equation (6) below.

Substituting equation (4) into equation (6) yields equation (7) below.

Also, the voltage equation on the q-axis of Equation (1) is represented by Equation (8) below.

Substituting equation (5) into equation (8) yields equation (9) below.

Here, FIG. 2 is a diagram showing the relationship between the three-phase dead time decreasing voltage and the motor currents Iu, Iv and Iw when the voltage across the switching element and the temperature are constant when the switching element is driven. The horizontal axis indicates the motor currents Iu, Iv and Iw, and the vertical axis indicates the three-phase dead time reduction voltage. Since the three-phase dead time reduction voltage characteristics are basically the same, FIG. 2 shows only one of the three phases.

As shown in FIG. 2, the dead time reduction voltage for negative motor currents Iu, Iv and Iw is the same magnitude as the dead time reduction voltage for positive motor currents Iu, Iv and Iw. , the sign is a negative value.

As shown in FIG. 2, the relationship between the three-phase dead time reduction voltage and the motor current has different characteristics depending on whether the magnitude (absolute value) of the motor current is greater than or equal to the threshold value Ith.

When the magnitude of the motor current is less than the threshold Ith, the dead time decreasing voltage is monotonically increasing. On the other hand, when the magnitude of the motor current is equal to or greater than the threshold value Ith, the dead time reduction voltage is substantially constant. Here, a first region in which the value of the motor current is equal to or greater than the threshold value Ith and a second region in which the value of the motor current is equal to or less than the threshold value −Ith are defined as saturation regions, and the value of the motor current is greater than the threshold value −Ith and less than the threshold value Ith. Let the third region be the unsaturated region. Note that the threshold Ith and the threshold −Ith are values that are appropriately determined in advance by experimental measurement or the like.

The reason why the relationship between the dead time decreasing voltage and the motor current is as follows is that when the magnitude of the motor current is less than the threshold value Ith, the dead times tdu, tdv, and tdw monotonically increase with respect to the magnitude of the motor current. On the other hand, when the magnitude of the motor current is equal to or greater than the threshold value Ith, the dead times tdu, tdv and tdw are substantially constant.

Here, monotonically increasing means that the value on the vertical axis increases continuously as the value on the horizontal axis increases. At this time, the rate of increase may not be constant. In the example of FIG. 2, monotonic increase means that the dead time reduction voltage continuously increases with increasing motor current value. Also in the following description, the term "monotonically increasing" is used with the same meaning.

In the first embodiment, it is assumed that each phase motor current when positioning the rotor 10 in steps S100 to S102 satisfies Condition 1 below.

Condition 1: When each phase motor current has a certain value (none of the values is 0), the magnitude (absolute value) of the dead time reduction voltage of each phase is the same.

Under this premise, none of the values of the motor currents in each phase is 0, so the d-axis phase θ_1 when positioning is given by (1), (3), (5), and (7) in FIG. ), (9) and (11).

Concerning the condition 1, a specific example is shown and described. When the values of the phase motor currents when the rotor 10 is fixed are Iu_1, Iv_1, and Iw_1, and these motor currents satisfy Condition 1, the magnitude of the dead time reduction voltage is the same, and the magnitude is ΔVc (>0).

There is a relationship of Iu_1+Iv_1+Iw_1=0 between the phase motor currents, and none of the values is 0, so the three signs of Iu_1, Iv_1 and Iw_1 are "two positive and one negative". Or "two negative, one positive".

Here, FIG. 5 shows the case of "Iu_1>0, Iv_1>0, Iw_1<0" as an example of "two are positive and one is negative". The dead time reduction voltage at motor currents Iu_1 and Iv_1 is ΔVc, and the dead time reduction voltage at Iw_1 is −ΔVc. That is, when the three-phase motor current values are Iu_1, Iv_1, and Iw_1, respectively, the magnitude (absolute value) of the dead time reduction voltage of each phase is ΔVc. Motor currents Iu_1, Iv_1 and Iw_1 may be in the saturated region or the non-saturated region. FIG. 5 shows the magnitudes (absolute value ) are the same.

Thus, on the premise that condition 1 is satisfied, the identifier 110 determines the three-phase motor currents as the first value (Iu_1), the second value (Iv_1), and the third value (Iw_1), respectively. ) to stop the rotor 10 of the motor 101 at the first position for identification. In the example of FIG. 5, the first value, the second value, and the third value may each be a value within the saturation region or a value within the non-saturation region. However, the identifier 110 determines that the three signs of the first, second, and third values are either two positive and one negative, or two negative and one positive. The first value, the second value, and the third value are set such that

In the example of "Iu_1>0, Iv_1>0, Iw_1<0", the d-axis phase θ_1 during positioning is within the range indicated by the arrow (3) in FIG.

Further, when the "relationship between each phase motor current (U-phase, V-phase, W-phase) and dead time reduction voltage" is the same for all three phases in the saturation region, any motor current within the saturation region satisfies Condition 1.

FIG. 6 shows each phase motor current Iu_1 when positioning the rotor 10 when the "relationship between the motor currents (U-phase, V-phase, W-phase) and the dead time reduction voltage" is the same for all three phases in the saturation region. , Iv_1 and Iw_1 are in the saturation region. In FIG. 6, the horizontal axis indicates the three-phase motor currents Iu, Iv and Iw, and the vertical axis indicates the three-phase dead time reduction voltage. In FIG. 6, a solid line 71 indicates the case of the U phase, a dashed line 72 indicates the case of the V phase, and a dotted line 73 indicates the case of the W phase.

In the case of FIG. 6, when each phase motor current is an arbitrary value within the saturation region (none of the values are 0), the magnitude (absolute value) of the dead time reduction voltage of each phase is , ΔVc and become the same. Therefore, the identifier 110 performs identification by setting the first value, the second value, and the third value to values within the saturation region. However, even in this case, there is a relationship of Iu_1+Iv_1+Iw_1=0 between the phase motor currents. Therefore, the identifier 110 determines that the three signs of the first, second, and third values are either two positive and one negative, or two negative and one positive. The first value, the second value, and the third value are set to values within the saturation region so that

It should be noted that the certain value or the certain region where all three phase dead-time reduction voltages have the same magnitude is often not known in advance. Therefore, the target current for motor current control may not be known, and the process may not proceed. However, in the example of FIG. 6, it is known that the region in which the dead time reduction voltages of the three phases all have the same magnitude is the saturation region. Since the absolute value of the motor current is in a range equal to or greater than the threshold Ith, the saturation region is a wide range and is a range determined by characteristic characteristics. Therefore, it is often known in advance which region of the motor current is the saturation region. Therefore, the example of FIG. 6 in which the three-phase motor current is set within the saturation region is realistic and suitable for practical use.

FIG. 7 shows the motor current Iu_1 when the d-axis phase θ_1 for positioning is the phases (1), (3), (5), (7), (9) and (11) in FIG. , Iv_1 and Iw_1, and the values of three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1 at that time. The d-axis dead-time reduction voltage ΔVd_1 and the q-axis dead-time reduction voltage ΔVq_1 are three-phase/dq-converted three-phase dead-time reduction voltages ΔVu_1, ΔVv_1, and ΔVw_1, and are given by the following equation (10). The formulas for calculating the d-axis dead time reduction voltage ΔVd_1 and the q-axis dead time reduction voltage ΔVq_1 in (1), (3), (5), (7), (9) and (11) of FIG. It can be obtained by substituting ΔVu_1, ΔVv_1, and ΔVw_1 in FIG. 7 into (10), which are shown in FIG. FIG. 8 calculates the d-axis dead time reduction voltage ΔVd_1 and the q-axis dead time reduction voltage ΔVq_1 in (1), (3), (5), (7), (9) and (11) of FIG. It is a table showing a calculation formula for.

FIG. 9 shows the results obtained by substituting ΔVd_1 and ΔVq_1 in FIG. 8 into the equations (7) and (9) and solving for the armature winding resistance Ra of the armature winding 4 and the dead time reduction voltage magnitude ΔVc. show. The armature winding resistance Ra of the armature winding 4 and the magnitude ΔVc of the dead time reduction voltage can be identified using the calculation formula of FIG. FIG. 9 shows the armature winding resistance Ra of the armature winding 4 and the dead time reduction in phases (1), (3), (5), (7), (9) and (11) of FIG. 4 is a table showing calculation formulas for identifying voltage magnitude ΔVc.

Further, FIG. 10 shows a result obtained by substituting the magnitude ΔVc of the dead time reduction voltage of FIG. 9 into the calculation formulas of the d-axis dead time reduction voltage ΔVd_1 and the q-axis dead time reduction voltage ΔVq_1 of FIG.

Specifically, FIG. 10 shows the d-axis dead time reduction voltage ΔVd_1 and the q-axis dead time reduction in (1), (3), (5), (7), (9) and (11) of FIG. 4 is a table showing a calculation formula for identifying voltage ΔVq_1;

Also, the three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1 are calculated by substituting the magnitude ΔVc of the dead time reduction voltage of FIG. 9 for the values of the three phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1 of FIG. formula can be obtained.

11 to 13 show formulas for calculating the three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1. Specifically, FIG. 11 shows calculations for identifying the U-phase dead time reduction voltage ΔVu_1 in (1), (3), (5), (7), (9) and (11) of FIG. It is the table|surface which showed the formula. FIG. 12 is a table showing calculation formulas for identifying the V-phase dead time reduction voltage ΔVv_1 in (1), (3), (5), (7), (9) and (11) of FIG. is. FIG. 13 is a table showing calculation formulas for identifying the W-phase dead time reduction voltage ΔVw_1 in (1), (3), (5), (7), (9) and (11) of FIG. is.

In step S104, the identifier 110 calculates the d-axis voltage command value Vd ^* _1, the q-axis voltage command value Vq ^* _1, the d-axis current Id_1, the q-axis current Iq_1, and the above equation (1) by the calculation method described above. Using the motor voltage equations, the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1 are calculated.

In step S105, when the identification of the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1 are completed, the process of the identifier 110 proceeds to step S106.

In step S106, the identifier 110 terminates the application of the d-axis current command value Id ^* _1 and the q-axis current command value Iq ^* _1 held to stop the rotor 10. FIG.

From the above, when each phase motor current (U-phase, V-phase, W-phase) is not zero, the three-phase dead time reduction voltage (points A, B and C in FIG. 5; points A and B in FIG. 6) at that motor current and C) can be identified.

Even if it is necessary to obtain the three-phase dead time reduction voltage with a magnitude other than the above motor current, if the condition 1 is satisfied, the three-phase dead time reduction voltage can be identified by the same processing as in FIG. .

If the armature winding resistance Ra of the armature winding 4 is known, only the method of identifying the three-phase dead time reduction voltage should be applied.

(Effect of Embodiment 1)
As described above, in Embodiment 1, even if the identifier 110 does not hold in advance the relationship between the three-phase dead time reduction voltage and the motor current of each phase as shown in FIG. From the values, the dq-axis currents, and the motor voltage equation, the three-phase dead-time reduction voltages can be identified. Therefore, it is not necessary to measure the relationship between the three-phase dead time reduction voltage and the motor current of each phase in advance, and the trouble of measuring can be saved. FIG. 2 is a diagram showing an example of the relationship between the three-phase dead time reduction voltage and the motor current. In FIG. 2, the horizontal axis indicates the motor current of each phase, and the vertical axis indicates the three-phase dead time reduction voltage. Note that FIG. 2 shows only one of the three phases. In Embodiment 1, it is not necessary to hold in advance the relationship between the three-phase dead time reduction voltage and each phase motor current as shown in FIG. An increase in dimensions can be suppressed.

Further, in Embodiment 1, even if there is an output voltage error and an armature winding resistance error of the permanent magnet synchronous motor 101, the three-phase dead time reduction voltage and the armature winding resistance are simultaneously and accurately identified. be able to. As a result, a decrease in responsiveness of the operation of the permanent magnet synchronous motor 101 and loss of synchronism can be avoided, and stable operation can be realized.

Further, in the identification, it is necessary to select a high frequency and a low frequency as the first and second switching frequencies in the technique described in Patent Document 1 above. Therefore, there are problems such as switching loss, heat generation, and noise associated therewith. Specifically, at high frequencies, there is a problem that switching loss increases and heat generation increases. Also, at lower frequencies the noise problem becomes more pronounced. In addition, since processing with two switching frequencies is required, there is also the problem that processing takes longer time than processing with one switching frequency. In contrast, in the first embodiment, multiple switching frequencies are not set when the d-axis current command value Id ^* _1 is given. Thus, in Embodiment 1, since only one switching frequency is selected, the problems of switching loss, heat generation, and noise can be avoided. Also, the processing time spent for identification can be shortened.

Further, in the identification, in the technique described in the above-mentioned Patent Document 1, in order to prevent the AC motor from rotating, there is a ) is applied with a direct current. Furthermore, the on-delay compensation value is calculated using the current control output acquired at that time. This method has the problem that the stop position of the rotor of the AC motor, at which the on-delay compensation value can be calculated, is limited to the position where the DC current flows between the two phases, and cannot be an arbitrary position. Therefore, in the technique of Patent Document 1, there is a problem that the on-delay compensation value cannot be calculated when the rotor of the AC motor must be stopped at the position where the DC current flows between the three phases. On the other hand, in Embodiment 1, if the motor current satisfies the condition 1, the rotor 10 of the permanent magnet synchronous motor 101 is fixed at the first position, and the three-phase dead time reduction voltage is identified. can do.

Embodiment 2.
Embodiment 2 describes an identification method different from the identification method of the identifier 110 described in Embodiment 1 above. Since the basic configurations and operations of the motor control device 100 and the permanent magnet synchronous motor 101 are the same as those of the first embodiment, description thereof will be omitted here, and the first embodiment will be referred to. The difference between the second embodiment and the first embodiment is the "relationship between each phase motor current and the dead time reduction voltage" when positioning the rotor 10, and the d-axis phase θe when positioning the rotor 10. , and the operation of the identifier 110 only.

In the first embodiment described above, the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* generated by the current controller 107, the d-axis current Id and the q-axis current generated by the three-phase/dq converter 104 Iq has been described as sampling only once. In contrast, in the second embodiment, an identification method will be described in which the stop position of the rotor 10 is changed stepwise to two positions and sampling is performed twice.

The two positions where the rotor 10 is stopped are indicated by the symbols (2), (4), (6), (8), (10) and (12) in FIG. It is a position that is one of six phases.

In the six phases (2), (4), (6), (8), (10) and (12) in FIG. 4, one of the three-phase motor currents Iu, Iv and Iw The phase motor current becomes zero, and the remaining two phase motor currents have the same motor current magnitude (absolute value) but opposite signs.

In the first sampling, the motor currents of two phases out of the three phases are set to the first value Ib_1 and the second value -Ib_1, respectively. Then, in the second sampling, the identifier 110 performs identification by setting the motor current of two phases out of the three phases to the fourth value Ib_2 and the fifth value -Ib_2, respectively.

Therefore, in the second embodiment, as the first sampling, the identifier 110 controls the motor current of one of the three phases of the permanent magnet synchronous motor 101 to the first value, and controls the remaining one phase to the first value. The rotor 10 is stopped at the first position by controlling to a second value having a value opposite in sign to the value of 1. It should be noted that although it is described here that the motor current is controlled, in practice, the dq-axis current may be controlled.

Next, the identifier 110 determines the d-axis voltage command value Vd ^* , the q-axis voltage command value Vq ^* , the d-axis current Id and the q-axis current Iq with the rotor 10 stopped at the first position. These are obtained as a first d-axis voltage command value Vd ^* _1, a first q-axis voltage command value Vq ^* _1, a first d-axis current Id_1 and a first q-axis current Iq_1, respectively.

Next, as the second sampling, the identifier 110 controls the motor current of one of the three phases of the permanent magnet synchronous motor 101 to a fourth value different in magnitude from the first value, and the remaining is controlled to a fifth value opposite in sign to the fourth value, the rotor 10 is stopped at the second position. The second location may be the same as the first location or may be different. Next, the identifier 110 determines the d-axis voltage command value Vd ^* , the q-axis voltage command value Vq ^* , the d-axis current Id and the q-axis current Iq with the rotor 10 stopped at the second position. These are obtained as a second d-axis voltage command value Vd ^* _2, a second q-axis voltage command value Vq ^* _2, a second d-axis current Id_2 and a second q-axis current Iq_2, respectively.

Next, the identifier 110 uses the first and second dq-axis voltage command values, the first and second dq-axis currents, and the motor voltage equation of the above equation (1) to determine the armature winding Identify the armature winding resistance Ra of line 4 and the three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1.

The operation of the identifier 110 according to the second embodiment will be described in detail below. FIG. 14 is a flowchart of automatic measurement processing of motor parameters and inverter parameters in the identifier 110 of the motor control device 100 according to the second embodiment. The process of FIG. 14 is performed by the identifier 110 as an application for obtaining the motor parameters of the permanent magnet synchronous motor 101 and the inverter parameters of the power converter 109 . The identification processing shown in FIG. 14 can be performed when a command to start the permanent magnet synchronous motor 101 is given. The permanent magnet synchronous motor 101 can be driven using the motor parameters and inverter parameters obtained by the identification process. The motor parameter identified by this identification process is the armature winding resistance Ra of the armature winding 4, and the inverter parameters are the three-phase dead time decreasing voltages ΔVu_1, ΔVv_1 and ΔVw_1. Alternatively, dead times tdu, tdv, tdw may be identified as inverter parameters.

When starting the processing shown in FIG. 14, the identifier 110 first performs a positioning operation of rotating and fixing the rotor 10 to a preset first position in steps S200 and S201. This operation is for obtaining the dq-axis voltage command values necessary for identifying the motor parameters and identifying the inverter parameters while the rotor 10 of the permanent magnet synchronous motor 101 is stationary.

To describe this operation in detail, first, in step S200, the identifier 110, in order to position the rotor 10, rotates the dq axis, which is the field axis of the rotating coordinate system of the permanent magnet synchronous motor 101, to the dq axis. A current command value and a position (d-axis phase θe) relative to the reference position of the rotating coordinate system are given. As the dq-axis current command value to be given, a preset d-axis current command Id ^* _1 (=DC value) is set on the d-axis, and a q-axis current command Iq ^* _1 (=0) is set on the q-axis. ). Also, θ_1 is set as the d-axis phase. During execution of the voltage command value acquisition process, the d-axis current command value Id ^* _1 and the q-axis current command value Iq ^* _1 (=0) are set so that the rotor 10 does not move from the first position. , and the d-axis phase θ_1. At this time, the d-axis current command value Id ^* _1 and the q-axis current command value Iq ^* _1 (=0) are set by the identifier 110, but are set by the speed controller 106. good too. Also, although the d-axis phase θ_1 is set by the identifier 110, it may be set by the estimator 105 as well.

Here, in the six phases indicated by symbols (2), (4), (6), (8), (10) and (12) in FIG. 4, the three-phase motor currents Iu, Iv and Iw Of these, the motor current of one phase is zero, and the motor currents of the remaining two phases have the same magnitude (absolute value) of the motor current and opposite signs.

In step S200, the identifier 110 selects one of the six phases indicated by (2), (4), (6), (8), (10) and (12) in FIG. Set the d-axis phase θ_1.

By the above processing, the convergence of the d-axis current and the q-axis current is completed in step S201, and the positioning of the rotor 10 is completed.

Next, in step S202, the identifier 110 acquires the d-axis voltage command value Vd ^* _1 and the q-axis voltage command value Vq ^* _1 from the current controller 107, and the d-axis current Id_1 from the three-phase/dq converter 104. and the q-axis current Iq_1. At this time, d-axis current command ^{value Id*_1 and q-axis current command value Iq* _1} may be obtained from speed controller 106 instead of d-axis current Id_1 and q-axis current Iq_1.

When the first positioning and sampling are completed in the processing of steps S200 to S202, the identifier 110 performs a positioning operation of rotating and fixing the rotor 10 to a preset second position in steps S203 and S204. I do. This operation is for obtaining the dq-axis voltage command values necessary for identifying the motor parameters and identifying the inverter parameters while the rotor 10 of the permanent magnet synchronous motor 101 is stationary.

Specifically, first, in step S203, the identifier 110 updates the d-axis current command value to a value different from that of the first positioning. As a current command to be given, a preset d-axis current command value Id ^* _2 (=DC value) is set on the d-axis, and a q-axis current command value Iq ^* _2 (=0), which is zero, is set on the q-axis. set. Also, θ_2 is set as the d-axis phase to be given. The d-axis phase θ_2 is one of the six phases indicated by symbols (2), (4), (6), (8), (10) and (12) in FIG.

Then, during the execution period of the dq-axis voltage command value acquisition process, the identifier 110 obtains the d-axis current command value Id ^* _2 and the q-axis current command value so that the rotor 10 does not move from the second position. Hold Iq ^* _2 (=0) and d-axis phase θ_2. At this time, the d-axis current command value Id ^* _2 and the q-axis current command value Iq ^* _2 (=0) are set by the identifier 110, but are set by the speed controller 106. good too. Also, although the d-axis phase θ_2 is set by the identifier 110, it may be set by the estimator 105 as well.

By the above processing, the convergence of the d-axis current and the q-axis current is completed in step S204, and the positioning of the rotor 10 is completed.

Next, in step S205, the identifier 110 acquires the d-axis voltage command value Vd ^* _2 and the q-axis voltage command value Vq ^* _2 from the current controller 107, and the d-axis current Id_2 from the three-phase/dq converter 104. and the q-axis current Iq_2. At this time, d-axis current command value Id ^* _2 and q-axis current command value Iq ^* _2 may be obtained from speed controller 106 instead of dq-axis currents Id_2 and Iq_2.

When the second positioning and sampling are completed in the processing of steps S203-S205, the processing of the identifier 110 proceeds to step S206.

In step S206, the identifier 110 starts identifying the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1. In the identification, as described below, the identifier 110 performs identification based on the voltage command value and the detected current acquired in step S202 and the voltage command value and detected current acquired in step S205. . Specifically, the identifier 110 outputs a d-axis voltage command value Vd ^* _1 and a q-axis voltage command value Vq ^* _1, a d-axis current Id_1 and a q-axis current Iq_1, a d-axis voltage command value Vd ^* _2 and a q-axis voltage Command value Vq ^* _2, d-axis current Id_2 and q-axis current Iq_2 are used.

Then, in step S207, the identifier 110 calculates the armature winding resistance Ra of the armature winding 4 and the dead time reduction voltages ΔVu_1, ΔVv_1 and ΔVw_1.

A motor voltage equation for the identification method of the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu_1, ΔV_1v and ΔVw_1, and "each phase motor current and dead time when positioning the rotor 10" Time Decrease Voltage” will be used for explanation. Since the motor voltage equation in the second embodiment is the same as the above equation (1) in the first embodiment, the above equation (1) is referred to.

Also, the actual motor applied voltages Vd_1 and Vq_1 on the dq coordinates at the time of the first sampling are applied to the d-axis voltage command value Vd ^* _1 and the q-axis voltage command value Vq ^* _1 on the dq coordinates after the d-axis dead time compensation. It is obtained by adding the voltage Vdcomp and the q-axis dead time compensation voltage Vqcomp and subtracting the d-axis dead time reduction voltage ΔVd_1 and the q-axis dead time reduction voltage ΔVq_1. This is the same as in the first embodiment. Therefore, the above equations (4) and (5) hold. Here, the re-statement of equations (4) and (5) is omitted, and the above equations (4) and (5) are referred to.

Let Vd_2 and Vq_2 be the voltages applied to the motor on the dq coordinates at the time of the second sampling, ΔVdcomp_2 and ΔVqcomp_2 be the dead time compensation voltages, and ΔVd_2 and ΔVq_2 be the dead time reduction voltages. Similar to Equation (5), Equations (11) and (12) below hold.

Note that the actual motor applied voltages Vd_2 and Vq_2 on the dq coordinates are the actual motor applied three-phase voltages Vu_2, Vv_2, and Vw_2 at the time of the second sampling, which are three-phase/dq-converted. Also, the d-axis dead time compensation voltage Vdcomp_2 and the q-axis dead time compensation voltage Vqcomp_2 on the dq coordinates are three-phase/dq-converted three-phase dead time compensation voltages Vucomp_2, Vvcomp_2, and Vwcomp_2 at the time of the second sampling. be.

In the second embodiment, the d-axis current command value Id ^* _1 and the q-axis current command value Iq ^* _1 are applied in the first sampling, and the d-axis current command values Id ^* _2 and q are applied in the second sampling. A shaft current command value Iq ^* _2 is applied. Since the rotor 10 is stationary in both the first sampling and the second sampling, the rotational angular velocity (electrical angle) ωe is ωe=0.

Therefore, at the time of the first sampling, the voltage equation on the d-axis and the voltage equation on the q-axis in the above equation (1) are respectively expressed by the above equations (7) and (9) as in the first embodiment. is represented by Here, the re-statement of equations (7) and (9) is omitted, and the above equations (7) and (9) are referred to.

Further, at the time of the second sampling, the voltage equation on the d axis and the voltage equation on the q axis in the above equation (1) are the following equations (13) and (14), respectively, as in the first sampling. is represented as

In the second embodiment, it is assumed that the phase motor currents for the first and second positioning of the rotor 10 in steps S200 to S205 satisfy Condition 2 below.

Condition 2: All three of the following (i) to (iii) must be satisfied.
(i): Of the phase motor currents, the motor current of one phase is zero, and the motor currents of the remaining two phases have the same magnitude (absolute value) and opposite signs. .
(ii): The magnitude of the two-phase motor current differs between the first and second positioning.
(iii): The magnitude (absolute value) of the dead time reduction voltage in this two-phase motor current is the same in the first and second positioning.

Under this premise, the motor current of one phase is zero. ), (6), (8), (10) and (12).

Concerning condition 2, a specific example is shown and described. The values of the phase motor currents for positioning the rotor 10 are Iu_1, Iv_1 and Iw_1 for the first positioning, and Iu_2, Iv_2 and Iw_2 for the second positioning.

When this motor current satisfies Condition 2, the magnitude (absolute value) of the dead time reduction voltage is the same in the phases where the motor current does not become zero during the first and second positioning, and the magnitude is ΔVb (>0).

Here, the phases in which the motor current becomes zero during positioning are the V phase in the first positioning and the W phase in the second positioning.

Regarding the magnitude (absolute value) of the motor current, the first positioning U-phase and W-phase are Ib_1, and the second positioning U-phase and V-phase are Ib_2.

among them,
"Iu_1=Ib_1, Iv_1=0, Iw_1=-Ib_1"
"Iu_2=Ib_2, Iv_2=-Ib_2, Iw_2=0"
is shown in FIG.

FIG. 15 is a diagram showing a case where the "relationship between motor currents (U-phase, V-phase, W-phase) and dead-time reduction voltage" does not completely match all three phases over the entire region. FIG. 15 shows that the motor currents Iu_1 and Iv_1 during the first positioning and the motor currents Iu_2 and Iw_2 during the second positioning have the same magnitude (absolute value) of the dead time reduction voltage, and the motor currents Iu_1 and Iv_1 during the first positioning are the same. It shows that the motor current Iw_1 at the time of the second positioning and the motor current Iv_2 at the time of the second positioning are zero. In FIG. 15, the horizontal axis indicates the three-phase motor currents Iu, Iv and Iw, and the vertical axis indicates the three-phase dead time reduction voltage. In FIG. 15, a solid line 81 indicates the case of the U phase, a broken line 82 indicates the case of the V phase, and another broken line 83 indicates the case of the W phase.

In the case of FIG. 15, the d-axis phase θ_1 and the d-axis phase θ_2 are phases indicated by symbols (2) and (12) in FIG. 4, respectively.

If the "relationship between each phase motor current (U-phase, V-phase, W-phase) and dead time reduction voltage" is the same for all three phases in the saturation region, any motor current in the saturation region satisfies Condition 2.

FIG. 16 is a diagram showing a case where the "relationship between the motor current (phases U, V, W) and the dead-time reduction voltage" matches all three phases (when the three phases match all regions, naturally saturation also match on the region). FIG. 16 shows that the motor currents Iu_1 and Iv_1 during the first positioning and the motor currents Iu_2 and Iw_2 during the second positioning are in the saturation region, and the motor current Iw_1 during the first positioning and the motor current Iw_1 during the second positioning are in the saturation region. is zero.

In FIG. 16, the horizontal axis indicates the three-phase motor currents Iu, Iv and Iw, and the vertical axis indicates the three-phase dead time reduction voltage. In FIG. 16, the U-phase case, the V-phase case, and the W-phase case overlap, so the same solid line 91 indicates the three-phase case.

Next, the relationship between the d-axis phases θ_1 and θ_2 will be described.

If there is no precondition that the "relationship between the motor current (U phase, V phase, W phase) and dead time reduction voltage" matches in the saturation region (for example, in the case of FIG. 15), the three phase dead time In order to identify all the decreasing voltages (U phase, V phase, W phase), the d-axis phase θ_1 and the d-axis phase θ_2 are different from each other (however, the phase difference between the d-axis phase θ_1 and the d-axis phase θ_2 is (except for π).

This is because when the d-axis phases θ_1 and θ_2 are the same or are shifted by π, only two of the three-phase dead time reduction voltages can be obtained. For example, when θ_1=π/6 and θ_2=π/6, ΔVv_1=0 and ΔVv_2=0 as shown in FIG. 17, and a V-phase dead time reduction voltage other than zero cannot be identified. FIG. 17 shows that the d-axis phase θ_1 in the first positioning and the d-axis phase θ_2 in the second positioning correspond to (2), (4), (6), (8), and (10) in FIG. ) and (12), and the values of the three-phase dead time reduction voltage at that time.

However, if the "relationship between the motor current (phases U, V, W) and the dead time reduction voltage" matches in the saturation region as shown in FIG. =0 and ΔVv_2=0, the V-phase dead time reduction voltage can be identified. The reason is that there is a precondition that the "relationship between the motor current (U phase, V phase, W phase) and the dead time reduction voltage" matches in the saturation region, so the U phase dead time reduction voltage or the W phase dead time reduction voltage can be used as the V-phase dead time decreasing voltage.

Therefore, if the "relationship between the motor currents (U-phase, V-phase, W-phase) and the dead time reduction voltage" matches in the saturation region, even if the d-axis phases θ_1 and θ_2 are the same phase, they are different phases. I don't mind.

In the following description, three-phase motor currents at the time of the first sampling are referred to as Iu_1, Iv_1, and Iw_1, and three-phase dead time reduction voltages at that time are referred to as ΔVu_1, ΔVv_1, and ΔVw_1. The dq-axis dead-time reduction voltages ΔVd_1 and ΔVq_1 are three-phase/dq-converted three-phase dead-time reduction voltages ΔVu_1, ΔVv_1, and ΔVw_1, and are given by the above equation (10). The three-phase motor currents at the second sampling are called Iu_2, Iv_2, and Iw_2, and the three-phase dead time reduction voltages at that time are ΔVu_2, ΔVv_2, and ΔVw_2. The dq-axis dead-time reduction voltages ΔVd_2 and ΔVq_2 are three-phase/dq-converted three-phase dead-time reduction voltages ΔVu_2, ΔVv_2, and ΔVw_2. is given by the following equation (15) with

Three-phase motor current Iu_1 at the first and second sampling times in the six phases indicated by symbols (2), (4), (6), (8), (10) and (12) in FIG. , Iv_1, Iw_1, Iu_2, Iv_2, and Iw_2, and three-phase dead time reduction voltages ΔVu_1, ΔVv_1, ΔVw_1, ΔVu_2, ΔVv_2, and ΔVw_2 are shown in FIG. Formulas (16), (17), (18) and (19) are given.

Substituting equation (16) into equation (7) and equation (17) into equation (13) yields equations (20) and (21).

Equations (22) and (23) are equations (22) and (23) obtained by solving equations (20) and (21) for Ra and ΔVb. Equations (22) and (23) are the armature winding resistances of the armature winding 4 in (2), (4), (6), (8), (10) and (12) of FIG. It is a calculation formula for calculating Ra and the magnitude ΔVb of the dead time reduction voltage.

The armature winding resistance Ra of the armature winding 4 and the magnitude ΔVb of the dead time reduction voltage can be identified by using the calculation formulas (22) and (23).

Furthermore, equations (24) and (25) are obtained by substituting the equation (23) for calculating ΔVb into the above equations (16) and (17).

Also, the values of ΔVq_1 and ΔVq_2 can be directly used in equations (18) and (19).

Equations (24), (25), and (18) and (19) are given by (2), (4), (6), (8), (10) and (12) in FIG. , dq-axis dead time reduction voltage values ΔVd_1, ΔVd_2 and ΔVq_1, ΔVq_2.

18 and 19 show the results obtained by substituting the values of the three-phase dead time reduction voltages ΔVu_1, ΔVv_1, ΔVw_1, ΔVu_2, ΔVv_2, and ΔVw_2 in FIG. . FIG. 18 shows the motor current when the d-axis phase θ_1 in the first positioning is (2), (4), (6), (8), (10) and (12) in FIG. 10 is a table showing values and three-phase dead time reduction voltage values at that time. Also, FIG. 19 shows the motor when the d-axis phase θ_2 in the second positioning is (2), (4), (6), (8), (10) and (12) in FIG. 4 is a table showing values of current and values of three-phase dead time reduction voltage at that time;

18 and 19 show the first and second samplings in the six phases labeled (2), (4), (6), (8), (10) and (12) in FIG. 3 shows calculation formulas for three-phase motor currents Iu_1, Iv_1, Iw_1, Iu_2, Iv_2, Iw_2 and three-phase dead time reduction voltages ΔVu_1, ΔVv_1, ΔVw_1, and ΔVu_2, ΔVv_2, ΔVw_2. By using the calculation formulas of FIGS. 18 and 19, the three-phase dead time reduction voltage can be identified.

In step S207 of FIG. 14, the identifier 110 performs identification using the dq-axis voltage command values and the motor current obtained by two samplings according to the calculation method described above. That is, identifier 110 uses dq-axis voltage command values Vd ^* _1, Vq ^* _1, Vd ^* _2 and Vq ^* _2, and dq-axis currents Id_1, Iq_1, Id_2 and Iq_2. Further, the identifier 110 identifies the armature winding resistance Ra of the armature winding 4 using the Ra calculation formula shown in Equation (22). Identifier 110 also calculates three-phase dead time reduction voltages ΔVu_1, ΔVv_1, ΔVw_1, and ΔVu_2, ΔVv_2, ΔVw_2 using the calculation formula shown in FIG.

At step S208 in FIG. 14, the identification of the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu_1, ΔVv_1, ΔVw_1, and ΔVu_2, ΔVv_2, ΔVw_2 is completed. In step S209, application of the d-axis current command value Id ^* _2 and the q-axis current command value Iq ^* _2 is terminated.

From the above, when each phase motor current (U-phase, V-phase, W-phase) is not zero, the dead time reduction voltage (points A, B, C and D in FIG. 15; points A and B in FIG. 16) at that motor current , C and D) can be identified.

After the above identification, an identification method (hereinafter referred to as identification method A) in the case where it is necessary to obtain the three-phase dead time reduction voltage with a magnitude other than the motor current will be described. In that case, the armature winding resistance Ra of the armature winding 4 has already been identified. If the armature winding resistance Ra of the armature winding 4 has already been identified, ΔVb can be calculated by either formula (20) or formula (21), that is, by one sampling. By substituting this calculation formula for ΔVb into formula (16) or formula (17) and ΔVb in FIG. 17, the dq-axis and three-phase dead time reduction voltages can be calculated. Therefore, the dq-axis and three-phase dead-time reduction voltages can be identified in one positioning.

Here, a method for identifying the three-phase dead-time reduction voltage in the non-saturation region using the identified armature winding resistance Ra, that is, the identification method A will be described.

As shown in FIG. 16, the "relationship between motor current (U phase, V phase, W phase) and dead time reduction voltage" is naturally non-uniform because the magnitude of dead time reduction voltage is the same in the entire range of motor current. They match even in the saturation region. In this case, in the non-saturation region, (iii) of Condition 2, "The magnitude (absolute value) of the dead time reduction voltage in this two-phase motor current is the same in the first and second positioning. ” is not satisfied. This is because the dead time reduction voltage in the non-saturation region monotonically increases, so the magnitude (absolute value) of the dead time reduction voltage in the motor current during the first and second positioning does not become the same.

Since all of the conditions 2 (i) to (iii) are not satisfied, the identification method shown in FIG. cannot identify the three-phase dead-time reduction voltage of

However, as shown in FIG. 16, in the "relationship between motor current (phases U, V, W) and dead time reduction voltage", when the motor current is in the saturation region, the magnitude of the dead time reduction voltage is the same. If so (because the entire region matches, it naturally matches even in the saturated region), it can be identified by the following method. First, the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltage in the saturation region are identified by the identification method shown in FIG. After that, using the identified Ra, the three-phase dead time reduction voltage in the non-saturation region is identified by the identification method A described above.

(Effect of Embodiment 2)
Also in the second embodiment, it is possible to obtain the same effect as in the first embodiment described above.

Furthermore, in the second embodiment, two (or one) phases (2), (4), (6), (8), (10) and (12) in FIG. 4 are selected, and the motor current is A method of identifying the armature winding resistance Ra of the armature winding 4 and the three-phase dead-time reduction voltage by sampling twice with at least two magnitude steps has been shown. As shown in FIG. 6 or FIG. 16, when the "relationship between the motor current (phases U, V, and W) and the dead time reduction voltage" is the same for all three phases in the saturation region, Thus, the armature winding resistance Ra of the armature winding 4 and the three-phase dead time decreasing voltages ΔVu, ΔVv and ΔVw can be identified by one sampling. However, as shown in FIG. 15, there are cases where the phase motor currents for the first and second positioning satisfy the condition 2 but not the condition 1. FIG. In this case, the identification method of performing only one sampling described in Embodiment 1 identifies the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu, ΔVv and ΔVw. may not be possible. In that case, if the identification method of sampling twice shown in the second embodiment is used, the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltages ΔVu, ΔVv and ΔVw can be identified. It can be performed.

As shown in FIG. 16, when the "relationship between the motor current (U phase, V phase, W phase) and the dead time reduction voltage" matches in the non-saturation region, the method described in Embodiment 1 is applied to the non-saturation region It is not possible to identify dead-time reduction voltages for three phases. However, as shown in FIG. 16, if the "relationship between motor current (phases U, V, W) and dead time reduction voltage" matches not only in the non-saturation region but also in the saturation region, the following method can be used for identification. can do. First, by the identification method shown in FIG. 14, the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltage in the saturation region are identified by positioning at least twice. Thereafter, using the identified Ra, the three-phase dead-time reduction voltage in the non-saturation region can be identified using identification method A above for at least one positioning.

Of course, it is also possible to identify the three-phase dead-time reduction voltage in the saturation region by using the identification method A described above, using the identified Ra, with at least one positioning.

(Modification of Embodiment 1 and Embodiment 2)
In the above-described second embodiment, the method of changing the motor current stepwise to at least two magnitudes and identifying the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltage has been described. . At that time, the cases in the six phases indicated by the symbols (2), (4), (6), (8), (10) and (12) in FIG. 4 have been described. However, it is not limited to this case. Even in the case of the phases within the ranges indicated by the six arrows of (1), (3), (5), (7), (9) and (11) in FIG. to identify the armature winding resistance Ra of the armature winding 4 and the three-phase dead time reduction voltage.

The identification method of performing only one sampling shown in Embodiment 1 is defined as the first identification method, and the identification method of performing two samplings shown in Embodiment 2 in a step-by-step manner is defined as the second identification method. A third identification method is the identification method in which only one sampling is performed as shown in the second embodiment. At this time, the first identification method is used for the phases (1), (3), (5), (7), (9) and (11) in FIG. Preset to use the second identification method for phases 4), (6), (8), (10) and (12), or to use the third identification method if Ra has already been identified. You can leave it. In that case, the identifier 110 stores in memory a table in which the identification method is set for each position of the rotor 10, refers to the table according to the position of the rotor 10, and selects the first identification method or An identification method is selected and used from the second identification method or the third identification method.

In the first and second embodiments, the U-phase position of the armature winding 4 is treated as the reference position of the rotor 10, that is, the reference position (θe=0) of the d-axis phase θe. However, it is not limited to this case. For example, when the reference position is shifted, the phase of θ in FIGS. 4, 7 to 13, and 17 to 19 may be shifted by the same amount as the shifted phase.

1 stator, 2 stator core, 3 teeth, 4 armature winding, 10 rotor, 11 rotor core, 12 permanent magnet, 20 switching element, 21 switching element, 22 connection point, 23 gate, 25 free wheel diode, 30 power supply, 31 capacitor, 100 motor controller, 101 permanent magnet synchronous motor, 102 current sensor, 102a current sensor, 102b current sensor, 102c current sensor, 104 three-phase/dq converter, 105 estimator, 106 speed controller, 107 current controller, 108 dq/three-phase converter, 109 power converter, 110 identifier, 120 vector controller, 130 controller.

Claims (14)

a current detector that detects a motor current flowing through the three-phase armature winding of the motor;
a vector controller that controls the actual rotational angular velocity of the motor by generating a voltage command value based on the target rotational angular velocity input from the outside and the motor current detected by the current detection unit;
a power converter that supplies power to the motor according to the voltage command value generated by the vector controller;
an identifier that identifies motor parameters of the motor and inverter parameters of the power converter;
The identifier is
The rotor of the motor is stopped at a first position by controlling the motor current of at least two of the three phases to a first value and a second value, respectively, and the rotor is stopped at the first position. 1, the voltage command value generated by the vector controller and the motor current detected by the current detection unit are acquired, and based on the voltage command value and the motor current, the A motor parameter and the inverter parameter are obtained ,
The identifier is
At the time of the first sampling, the rotor of the motor is moved to the first position by controlling the motor current of at least two of the three phases to the first value and the second value, respectively. and the rotor is stopped at the first position, the voltage command value generated by the vector controller and the motor current detected by the current detection unit are obtained, With a voltage command value of 1 and a first motor current,
controlling the motor current of at least two phases out of the three phases to a fourth value and a fifth value different from the first value and the second value, respectively, at the time of the second sampling; The rotor of the motor is stopped at a second position, and with the rotor stopped at the second position, the voltage command value and the motor current are obtained, Let the voltage command value and the second motor current be
determining the motor parameter and the inverter parameter based on the first and second voltage command values and the first and second motor currents,
By controlling the motor current of at least two of the three phases to the first value and the second value, respectively, the rotor of the motor is stopped at the first position for identification. The first identification method is the method of
stopping the rotor of the motor at the first position by controlling the motor current of at least two of the three phases to the first value and the second value, respectively; By stepwise controlling the motor current of at least two of the three phases to the fourth value and the fifth value different from the first value and the second value, respectively When the method of identifying by stopping the rotor of the motor at the second position is the second identification method,
The identifier is
An identification method is preset for each position of the rotor from among the first identification method and the second identification method, and identification is performed by changing the identification method according to the position at which the rotor is stopped. I do,
motor controller. the motor parameter is an armature winding resistance of the armature winding of the motor;
the inverter parameter is the dead time reduction voltage of the motor;
The identifier is
having motor voltage information defining the relationship between the actual voltage of the motor, the motor current, and the armature winding resistance;
The rotor of the motor is stopped at the first position by controlling the motor current of at least two of the three phases to the first value and the second value, respectively, and the rotation is performed. With the motor stopped at the first position, the voltage command value generated by the vector controller and the motor current detected by the current detection unit are acquired, and the actual voltage of the motor is By subtracting the dead time reduction voltage from the voltage command value and adding the dead time compensation voltage corresponding to the dead time reduction voltage identified by the identifier, the voltage command value and the determining the armature winding resistance and the dead time reduction voltage using the motor current and the motor voltage information;
The motor control device according to claim 1. the first position of the rotor is a position where none of the motor currents of the three phases is zero;
3. A motor control device according to claim 2. When the motor currents of the three phases are respectively the first value, the second value, and the third value, the magnitudes of the dead time reduction voltages of the three phases are all the same. in the case of
The identifier is
The rotor of the motor is stopped at the first position by controlling the three-phase motor current to the first value, the second value, and the third value, respectively. and
The identifier is
of the signs of the first value, the second value, and the third value, such that two are positive and one is negative, or two are negative and one is positive, setting the first value, the second value and the third value;
4. A motor control device according to claim 3. When the value of the motor current is positive or negative and the absolute value of the motor current is equal to or greater than a threshold value as the saturation region,
When the motor currents of the three phases are respectively the first value, the second value, and the third value within the saturation region, the magnitude of the dead time reduction voltage of the three phases are all the same ,
The identifier is
Of the signs of the first value, the second value and the third value, two are positive and one is negative, or two are negative and one is positive. setting a value of 1, the second value, and the third value within the saturation region of the motor current;
5. A motor control device according to claim 4. the motor parameter is an armature winding resistance of the armature winding of the motor;
wherein the inverter parameter is the dead time reduction voltage of the motor;
A motor control device according to claim 5 . the first position and the second position of the rotor are positions at which one of the three phases of the motor current is zero;
7. A motor control device according to claim 6 . If the following condition 2 holds,
Condition 2: satisfying all three of the following (i) to (iii);
(i): Of the phase motor currents, one phase motor current is zero, and the remaining two phase motor currents are the same in magnitude and opposite in sign.
(ii): the magnitudes of the remaining two-phase motor currents are different between the first sampling time and the second sampling time;
(iii): the magnitude of the dead time reduction voltage in the remaining two phases of the motor current is the same between the first sampling and the second sampling;
The identifier is
Among the three phases of the motor current, the motor current of one phase is zero, and the absolute values of the motor currents of the remaining two phases are the same and the signs are opposite.
The first value and the fourth value are set at two different points in the region where the motor current is positive, and the second value and the fifth value are set at the regions where the motor current is negative. set at each of two different points to identify the armature winding resistance and the dead time reduction voltage;
The motor control device according to claim 7 . the d-axis phase of the motor current during the first sampling and the d-axis phase of the motor current during the second sampling are different from each other,
a phase difference between the d-axis phase of the motor current at the first sampling and the d-axis phase of the motor current at the second sampling is not π;
A motor control device according to claim 8 . When the value of the motor current is positive or negative and the region in which the absolute value of the motor current is equal to or greater than a threshold value is defined as a saturation region, in the saturation region of the motor current, the value of the dead time reduction voltage of the three phases is if all match,
The identifier is
Among the saturation regions, a region in which the value of the motor current is positive and the absolute value of the motor current is equal to or greater than a threshold is defined as a first region, and the value of the motor current is negative and the absolute value of the motor current is equal to or greater than the threshold. When the area of is the second area,
Among the three-phase motor currents, the first motor current is set so that the motor current of one phase is zero and the absolute values of the motor currents of the remaining two phases are the same and the signs are opposite. setting a value within the first region and setting the second value within the second region;
Among the three phases of the motor current, the motor current of one phase is zero, and the absolute values of the motor currents of the remaining two phases are the same and the signs are opposite. setting a value within the first region and setting the fifth value within the second region;
identifying the armature winding resistance and the dead time reduction voltage;
A motor control device according to claim 8 . The d-axis phase of the motor current at the first sampling and the d-axis phase of the motor current at the second sampling are the same phase or different phases,
11. A motor control device according to claim 10 . After identifying the armature winding resistance or when the armature winding resistance is known, the identifier determines, based on the armature winding resistance, identifying the dead time reduction voltage;
A motor control device according to any one of claims 2 to 11 . Utilizing the relationship that the actual voltage of the motor is obtained by subtracting the dead time reduction voltage from the voltage command value and adding the dead time compensation voltage corresponding to the dead time reduction voltage identified by the identifier. Then, using the armature winding resistance, the voltage command value, the motor current, and the motor voltage information, determine the dead time reduction voltage,
13. A motor control device according to claim 12 . When the value of the motor current is positive or negative and the region in which the absolute value of the motor current is less than a threshold value is defined as a non-saturation region,
The identifier is
After identifying the armature winding resistance by performing the first sampling and the second sampling, three-phase dead time reduction voltage in the non-saturation region based on the identified armature winding resistance to identify
The motor control device according to any one of claims 10-13 .

Images