JP2002051580A - Position-sensorless control method for synchronous motor, and position-sensorless controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a sensorless control method and a controller for a synchronous motor, which can materialize the smooth switchover of the angle estimation systems for low speed range and high speed range, in an estimation system which converges the error between the estimated angle of a rotor and the actual angle to zero. SOLUTION: An error for low speed by the estimation system for low speed is made, and an error for high speed by the estimation method for high speed is made, and an error after addition is made by substantially adding the error for low speed and the error for high speed at a specified rate, and the estimated error is corrected so that this error after addition may converge to zero, and the angle estimation system for low speed range or the high speed range is switched over smoothly.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a synchronous motor position sensorless control method and control apparatus for controlling a synchronous motor without using a position sensor. In particular, the present invention relates to a position sensorless control method and a control device for a synchronous motor capable of smoothly switching an angle estimation method between a low speed region and a high speed region.

[0002]

2. Description of the Related Art A synchronous motor having no mechanical commutation mechanism needs to flow a phase current in synchronization with the angle of the rotor at that time. In a conventional motor control device, a motor is controlled by obtaining information on the angle of a rotor using a position sensor such as a Hall element, a resolver, a magnetic encoder, or an optical encoder attached to a synchronous motor. In such a motor control device, it is necessary to provide a position sensor for the synchronous motor, and there is a problem that the cost increases by the position sensor and the size of the synchronous motor increases. As a device that solves such a problem, there has been a motor control device that achieves cost reduction and size reduction by eliminating position sensors and performing position sensorless control. As a typical example of such a position sensorless control device,
There are those described in the following four documents. (1) Conference Proceedings of Industry Applicati
ons Society (1998, 1998 IEEE) pages 671 to 67
Motor control device described on page 6 (hereinafter abbreviated as Conventional Example 1) (2) Motor control device disclosed in Japanese Patent Publication No. 2858692 (hereinafter abbreviated as Conventional Example 2) (3) JP-A-10-323099 Motor control device disclosed in the official gazette (hereinafter abbreviated as Conventional Example 3) (4) Transactions of the Institute of Electrical Engineers of Japan D117 Vol.
Motor control devices described on pages 104 to 104 (hereinafter abbreviated as Conventional Example 4) Hereinafter, Conventional Examples 1 to 4 will be described.

First, a description will be given of a position sensorless control device for a synchronous motor according to Conventional Example 1. The position sensorless control device of Conventional Example 1 controls the synchronous reluctance motor by adding a certain ratio between the estimated angle obtained by the low-speed estimation method and the estimated angle obtained by the high-speed estimation method. I have. In a synchronous reluctance motor, the inductance of the motor d-axis and 90
イ ン ダ ク タ ン ス The inductance of the advanced q axis is different. In the low-speed estimation method, the difference between the d-axis inductance and the q-axis inductance is used. That is, in a low speed range, a voltage pulse is applied, a current response to the voltage pulse is detected, and an estimated angle is created based on the current response.

In a synchronous reluctance motor, a magnetic flux rotates in synchronization with a rotor in a certain operation state, and a phase (magnetic flux phase) with respect to a rotational coordinate is constant. The high-speed estimation method utilizes the relationship between the magnetic flux and the phase. That is, in the high-speed range, the α-axis component and the β-axis component of the magnetic flux are calculated, the inverse tangent of the ratio between them is obtained, and the magnetic flux phase is subtracted to obtain the estimated angle. Further, in the switching range between the low-speed range and the high-speed range, the low-speed estimation angle is created by the low-speed estimation method, and the high-speed estimation angle is created by the high-speed estimation method. Then, the low-speed estimated angle and the high-speed estimated angle are added at a certain ratio. The ratio in the addition processing is changed according to the estimated speed.
That is, when the estimated speed changes from low speed to high speed, the ratio of the high speed estimated angle is set to gradually increase. In this way, the angle estimation method between the low-speed range and the high-speed range has been smoothly switched.

Next, a description will be given of a position sensorless control device for a synchronous motor according to Conventional Example 2. The position sensorless control device of Conventional Example 2 controls a permanent magnet synchronous motor by adding a speed command value and an estimated speed at a certain ratio. In the position sensorless control device of the second conventional example, first, forced synchronization is performed at the time of startup. Here, the forced synchronization means that a certain speed command value is created, and an alternating current having a frequency is supplied to the stator winding according to the speed command value. In the high-speed range, the estimated speed is created using the back electromotive force generated by the permanent magnet. The method of creating the estimated speed is almost the same as the method of creating a fourth conventional example described later. Further, after the start, the speed command value and the estimated speed are added at a certain ratio. In this addition process, the ratio of the estimated speed is gradually increased. And
After a lapse of a predetermined time, the estimated angle and the estimated speed are created only from the high-speed estimation method that estimates using the back electromotive force.

Next, a description will be given of a position sensorless control device for a synchronous motor according to a third conventional example. The position sensorless control device according to the third embodiment obtains an error at a low speed, corrects the estimated angle so that the error converges to 0, and controls the embedded magnet type synchronous motor. In an interior permanent magnet synchronous motor, the d-axis inductance and the q-axis inductance are different. In the position sensorless control device of Conventional Example 3, the difference between the d-axis inductance and the q-axis inductance is used. That is, an AC is superimposed on the d-axis current command value, and a q-axis current response to the AC is detected. Then, the estimated angle is corrected so that the q-axis current response converges to zero.

Next, a description will be given of a position sensorless control device for a synchronous motor according to a conventional example 4. The position sensorless control device of Conventional Example 4 calculates an error at high speed and corrects the estimated angle so that the error converges to zero. With this correction, the position sensorless control device of Conventional Example 4 controls the permanent magnet synchronous motor. When an angle error exists in a permanent magnet synchronous motor, an error occurs between the model current calculated using the motor constant and the actual current. Conventional example 4
The position sensorless control device uses an error between the model current and the actual current. That is, the estimated angle is corrected so that the error (γ-axis current error) between the actual current and the model current on the γ-axis converges to zero. Also, so that the error between the actual current on the δ-axis and the model current (δ-axis current error) converges to 0,
Correct the estimated back electromotive force.

[0008]

However, the conventional synchronous motor position sensorless control device described above has the following problems. The position sensorless control devices of Conventional Example 3 and Conventional Example 4 do not consider switching of the angle estimation method between the low-speed region and the high-speed region, and therefore perform smooth switching of the angle estimation method between the low-speed region and the high-speed region. Could not.

Further, the position sensorless control device of the prior art 2 smoothly switches from the forced synchronization to the high-speed estimation method, but performs the forced synchronization without performing the angle estimation in a low-speed region. High output torque could not be realized. Further, the position sensorless control device of Conventional Example 1 can smoothly switch the angle estimation method between the low-speed range and the high-speed range, but requires a long time for the calculation process because it performs the arc tangent calculation process. there were. Conventional Example 1
The position sensorless control device is mainly adapted to a synchronous reluctance motor, and is difficult to adapt to a permanent magnet synchronous motor.

[0010] In the above prior art, conventional example 1
Switching the angle estimation method between the low-speed range and the high-speed range in the conventional example 3
And a combination suitable for the conventional example 4. That is, in the low-speed range, an estimation angle is created using the estimation method of Conventional Example 3 as the low-speed estimation method. In a high-speed region, an estimation angle is created using the estimation method of Conventional Example 4 as an estimation method for high-speed use. Then, in the switching range between the low-speed range and the high-speed range, the low-speed estimation angle is created by the low-speed estimation method, and the high-speed estimation angle is created by the high-speed estimation method. Then, the low-speed estimated angle and the high-speed estimated angle are added at a certain ratio. In this addition processing, the addition ratio is changed according to the estimated speed at that time. That is, when the estimated speed changes from the low speed to the high speed, the ratio of the high speed estimated angle is set to gradually increase. However, in the estimation method based on the combination of Conventional Example 1, Conventional Example 3, and Conventional Example 4, smooth switching of the angle estimation method between the low-speed region and the high-speed region cannot be realized for the following reasons. .

In the switching range, as in the following equation (1):
The low-speed estimated angle θm (low speed), which is the estimated angle obtained by the low-speed estimation method, is set to θl. On the other hand, as shown in the following equation (2), the high-speed estimated angle θm (high-speed), which is the estimated angle obtained by the high-speed estimation method, is set to θh.
Note that KR indicates an addition ratio. Then, as shown in the following equation (3), the estimated angle θm (switching) is obtained by adding the estimated low-speed angle θl and the estimated high-speed angle θh at a certain ratio represented by (1-KR) and KR, respectively. ). Here, when the estimated speed changes from the low speed to the high speed, the ratio of the high speed estimated angle is gradually increased. That is,
The addition ratio KR is gradually increased.

Θl = θm (low speed) (1) θh = θm (high speed) (2) θm (switching) = (1−KR) · θ1 + KR · θh (0 <KR <1)・ (3)

By the way, parameter error, current sensor error, influence of dead time, difference between current command value and actual current,
The estimated angle deviates from the actual angle due to a difference between the voltage command value and the actual voltage, a calculation delay, and the like. This shift has a different value depending on the estimation method used. next,
Let's take an example. When the estimated angle is corrected so that the error (q-axis current response in Conventional Example 3) becomes 0 using only the low-speed estimation method as in the following equation (4), the estimated angle becomes θ10. Also, as shown in the following equation (5), when the estimated angle is corrected using only the high-speed estimation method so that the error (γ-axis current error and δ-axis current error in Conventional Example 4) becomes 0, the estimated angle is corrected. Becomes θh0. These estimated angles θl0 and θh0 are different values.

Θ10 = θm (low speed) (4) θh0 = θm (high speed) (5)

In such a case, the estimated angle θm (switching) obtained by Expression (3) is always different from the angles θ10 and θh0 expressed by Expressions (4) and (5). Therefore, the error does not always become 0, and the low-speed estimation method and the high-speed estimation method always keep correcting the estimated angles θl and θh. As a result, the deviation between the estimated angles θl and θh continues to increase, and may eventually lose synchronism. As described above, in the estimation method that converges the error to 0, the method of adding the angles at a certain rate in the switching region cannot realize a smooth switching between the angle estimation methods of the low speed region and the high speed region.
The present invention solves the above-described problem, and realizes a smooth switching between an angle estimation method between a low-speed region and a high-speed region in an estimation method that converges an error between an estimated angle and an actual angle to zero. It is an object of the present invention to provide a synchronous motor position sensorless control method and a control device that can perform the control.

[0016]

In order to achieve the above object, a position sensorless control method for a synchronous motor according to the present invention estimates a rotor angle in a synchronous motor to form an estimated angle, and forms the estimated angle. A position sensorless control method for controlling the synchronous motor based on: a step of periodically creating a first error that changes in accordance with an estimation error of the estimated angle formed using a first estimation method; ,
Periodically generating a second error that changes in accordance with the estimation error of the estimated angle formed using a second estimation method; and determining the first error and the second error by a predetermined value. A step of substantially adding at a ratio to create an error after addition; and a step of correcting the estimated angle so that the error after addition converges to zero. According to the present invention having such steps, it is possible to realize a position sensorless control method for a synchronous motor that smoothly switches the angle estimation method between a low speed region and a high speed region.

Further, in the synchronous sensor position sensorless control method according to the present invention, the step of creating a temporal change in the composite ratio and the step of synchronizing with the cycle in which the second error is created are performed. Changing the combination ratio based on According to the present invention having such steps, a position sensorless control method of a synchronous motor that can perform a smooth changeover of the angle estimation method between the low speed range and the high speed range with a smaller amount of calculation is realized.

The position sensorless control device for a synchronous motor according to the present invention includes an estimated angle creating means for creating an estimated angle of the rotor in the synchronous motor, and a driving means for driving the synchronous motor based on the estimated angle. A synchronous motor position sensorless control device, wherein the estimated angle creating means periodically creates a first error that changes according to an estimated error of the estimated angle formed using a first estimation method. A first error creating unit that periodically creates a second error that changes in accordance with the estimation error of the estimated angle formed using a second estimation method; The first error and the second error are substantially added by changing a combination ratio that is an addition ratio of the first error and the second error according to the number of rotations of the rotor. Create error after addition An error preparing means after calculation, the addition after the error is and a estimated angle correcting means for correcting the estimated angle to converge to zero. According to the present invention configured as described above, a position sensorless control device for a synchronous motor that can smoothly switch the angle estimation method between a low speed region and a high speed region is realized.

Further, the position sensorless control device for a synchronous motor according to the present invention includes a change amount creating means for creating a time-dependent change amount of the combined ratio, and an operation cycle of the second error creating means. A combination ratio changing unit that changes the combination ratio based on the change amount. According to the present invention configured as described above, a position sensorless control device for a synchronous motor that can smoothly switch the angle estimation method between the low speed region and the high speed region with a small amount of calculation is realized.

[0020]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a synchronous motor position sensorless control device according to an embodiment of the present invention.

Embodiment 1 Hereinafter, a position sensorless control device for a synchronous motor according to Embodiment 1 will be described. Example 1
In the synchronous motor position sensorless control device, the estimated angle is adjusted so that the result obtained by adding the angle error obtained by the low-speed estimation method and the angle error obtained by the high-speed estimation method at a certain rate converges to 0. to correct. By correcting the estimated angle in this manner, the position sensorless control device of the first embodiment smoothly switches the angle estimation method between the low speed range and the high speed range.

[Configuration of Position Sensorless Control Device of First Embodiment] First, the configuration of the position sensorless control device of the first embodiment will be described. FIG. 1 is a block diagram illustrating a configuration of a position sensorless control device for a synchronous motor according to the first embodiment. IPMSM (Interior Permanent Magnet Sy
nchronous Motor: embedded magnet type synchronous motor)
A stator (not shown) made of an electromagnetic steel sheet, and stator windings 11u, 11v, and 11w each made of a coated copper wire and wound around a stator (not shown) made of a coated copper wire.
And a rotor 1 which is disposed in close proximity to the stator.
2 are provided. Here, the stator winding 11u,
11v and 11w are Y connection (each stator winding 11u, 11w)
v, 11w are connected at one point). The rotor 12 includes a rotor yoke 13 made of an electromagnetic steel plate, a permanent magnet 14 disposed inside the rotor yoke 13, and a shaft 15 having the same rotation center as the rotor yoke 13. This rotor 12
Are rotatably supported, and are configured to rotate by the interaction between the magnetic flux generated by the phase current and the magnetic flux by the permanent magnet 14.

The synchronous motor position sensorless control device according to the first embodiment includes a current sensor 21u that outputs an analog u-phase current value ua and a current sensor 21v that outputs an analog v-phase current value iva. Each output signal is input to a microcomputer or a microprocessor (hereinafter abbreviated as a microcomputer) 22. Also, the microcomputer 22 has an analog u-phase current value iu
a, analog v-phase current value iva, and analog speed command value ω
* A is input and the switching command signals guh, gu
l, gvh, gvl, gwh, and gwl are output to the drive unit 30. The drive unit 30 outputs the switching command signals guh, g
ul, gvh, gvl, gwh, and gwl are input to control voltages applied to the stator windings 11u, 11v, and 11w.

[Configuration of Driving Unit 30] FIG. 2 is a circuit diagram showing a configuration of the driving unit 30 in the first embodiment. As shown in FIG. 2, the drive unit 30 includes a power supply 31, a collector connected to the positive electrode of the power supply 31, an emitter connected to the stator winding 11u,
11v, 11w connected to the upper IGBT (In
sulated Gate Bipolar Transistor 32u, 32v, 32w and upper flywheel diode 33 connected in anti-parallel to upper IGBTs 32u, 32v, 32w, respectively.
u, 33v, and 33w. The driving unit 30
Means that the collector of each transistor is a stator winding 11u,
11v and 11w, respectively, and the emitter is a power supply 3
Lower IGBTs 34u, 3 connected to the negative electrode of
4v, 34w and these lower IGBTs 34u, 34
v and 34w, respectively, and have lower flywheel diodes 35u, 35v and 35w connected in anti-parallel to each other, and a pre-drive device 36. The pre-drive unit 36 receives the switching command signal gu sent from the microcomputer 22.
The gate voltages of the upper IGBTs 32u, 32v, 32w and the gate voltages of the lower IGBTs 34u, 34v, 34w are controlled based on h, gul, gvh, gvl, gwh, gwl, respectively.

[Configuration of Microcomputer 22]
This is a generally used microcomputer composed of a CPU, a ROM, a RAM, a timer, an input / output port, and a bus connecting these. Microcomputer 2
2 includes a speed control unit 40, a current control unit 50, and an angle estimation unit 60 functionally. The speed controller 40 receives the analog speed command value ω * a and the estimated speed ωm,
It outputs a specified γ-axis current value iγ * and a specified δ-axis current value iδ *. The current control unit 50 controls the analog u-phase current value ua, the analog v-phase current value iva, the γ-axis current command value iγ *, the δ-axis current command value iδ *, the estimated angle θm, the estimated speed ωm, and the superimposed wave current command value iγa *. Is input, and the δ-axis current value iδ
, U-phase current value iu, v-phase current value iv, and u-phase voltage command value v
u *, v-phase voltage command value vv *, w-phase voltage command value vw *, and switching command signals guh, gul, gvh, gv
l, gwh, and gwl are output. The angle estimation unit 60 calculates δ
The shaft current value iδ, the u-phase current value iu, the v-phase current value iv, the u-phase voltage command value vu *, the v-phase voltage command value vv *, and the w-phase voltage command value vw * are input, and the estimated angle θm and Estimated speed ω
m and the superimposed wave current command value iγa * are output.

[Structure of Speed Control Unit 40] FIG. 3 is a block diagram showing the structure of the speed control unit 40 in the first embodiment. The speed control unit 40 includes an ADC (Analog Digital Converter).
ter: analog / digital converter) 41 and PI
It is composed of a control unit 42 and a torque / current conversion unit 43. The ADC 41 receives the analog speed command value ω * a and outputs a speed command value ω *. The PI control unit 42 receives the speed command value ω * and the estimated speed ωm, and outputs a torque command value T *. The torque / current converter 43 receives the torque command value T * and outputs a γ-axis current command value iγ * and a δ-axis current command value iδ *.

[Configuration of Current Control Unit 50] FIG. 4 is a block diagram showing the configuration of the current control unit 50 in the first embodiment. The current control unit 50 includes two ADCs 51u and 51v.
A three-phase to two-phase converter 52, a voltage command value generator 53,
It comprises a two-phase / three-phase converter 54 and a PWM controller 55. The ADC 51u receives the analog u-phase current value ua and outputs a u-phase current value iu, and the ADC 51v receives the analog v-phase current value iva and outputs a v-phase current value iv. The three-phase-to-two-phase converter 52 converts the u-phase current values iu and v
The phase current value iv and the thrust angle θm are input, and the γ-axis current value iγ and the δ-axis current value iδ are output. The voltage command value creation unit 53 receives the γ-axis current value iγ, the δ-axis current value iδ, the γ-axis current command value iγ *, the δ-axis current command value iδ *, the estimated speed ωm, and the superimposed wave current command value iγa *. Then, it outputs the specified γ-axis voltage value vγ * and the specified δ-axis voltage value vδ *. The two-phase / three-phase converter 54 receives the γ-axis voltage command value vγ *, the δ-axis voltage command value vδ *, and the estimated angle θm, and receives the u-phase voltage command value v
u *, v-phase voltage command value vv *, and w-phase voltage command value vw * are output. The PWM controller 55 has a u-phase voltage command value vu *
, The v-phase voltage command value vv * and the w-phase voltage command value vw *, and the switching command signals guh, gul, gv
Output h, gvl, gwh, gwl.

[Configuration of Angle Estimating Unit 60] FIG. 5 is a block diagram showing the configuration of the angle estimating unit 60 in the first embodiment. The low-speed estimating unit 61 receives the δ-axis current value iδ and the combined ratio α and outputs a low-speed error εl.
And an estimated high-speed operation unit 70. The estimated high-speed operation unit 70 receives the low-speed error ε1, the u-phase voltage command value vu *, the v-phase voltage command value vv *, the w-phase voltage command value vw *, the u-phase current value iu, and the v-phase current value iv. And the combined ratio α, the superimposed wave current command value iγa *, the estimated angle θm, and the estimated speed ω
and m. The estimation high-speed operation unit 70 includes a combination ratio changing unit 71, a superimposed wave creation unit 72, a high-speed estimation unit 73, and an angle / speed creation unit 74.

The combining ratio changing section 71 receives the estimated angle ωm and outputs a combining ratio α. The superimposed wave generator 72 receives the composite ratio α and outputs a superimposed wave current command value iγa *. The high-speed estimating unit 73 calculates the u-phase voltage command value vu *, the v-phase voltage command value vv *, the w-phase voltage command value vw *, and the u-phase current value i.
The u and v-phase current values iv, the estimated angle θm, and the estimated speed ωm are input, and a high-speed error εh is output. The angle / speed generating unit 74 calculates the combination ratio α, the error εl for low speed, and the error εh for high speed.
And outputs the estimated angle θm and the estimated speed ωm.

[Coordinate System] Next, the coordinate system of the synchronous motor in the first embodiment will be described. FIG. 6 is an explanatory diagram of the coordinate system according to the first embodiment. In FIG. 6, for simplicity of explanation, IP having two permanent magnets has two magnetic poles.
MSM is shown. The d-axis and the q-axis shown in FIG. The d-axis is in the same direction as the magnetic flux by the permanent magnet 14 disposed on the rotor 12,
The q axis is a direction advanced by 90 ° with respect to the d axis. And
The angle between the stator winding 11u and the d-axis is the angle θ. In FIG. 6, the rotation in the counterclockwise direction is defined as normal rotation. When the rotor 12 rotates forward, the angle θ advances. The direction of this forward rotation is the u-phase, v-phase, and w-phase stator windings 11u, 1
The currents flowing through 1v and 11w change in the order of u-phase, v-phase, and w-phase. The γ axis and the δ axis are estimated angles θm
Is the axis defined by In FIG. 6, the axis rotated by the estimated angle θm from the stator winding 11u is the γ axis, and the δ axis is a direction advanced by 90 ° with respect to the γ axis. Further, the difference between the angle θ and the estimated angle θm is calculated as an angle error Δθ (= θ
−θm).

FIG. 6 shows a case where there is a positive angle error Δθ. Here, there is no error in the angle estimation and the angle error Δ
When θ is 0, the estimated angle θm matches the angle θ, the d-axis matches the γ-axis, and the q-axis matches the δ-axis. In the following description, the angle θ, the estimated angle θm, and the angle error Δθ are represented by electrical angles. Hereinafter, unless otherwise specified, values relating to angles are represented by electrical angles. Here, the mechanical angle is the rotor 12
The electrical angle and the mechanical angle are (electrical angle)
= (P / 2) · (mechanical angle). Here, p is the number of magnetic poles.

[Operation of the Synchronous Motor Position Sensorless Control Device of the First Embodiment] Next, the operation of the synchronous motor position sensorless control device of the first embodiment will be described. The analog speed command value ω * a generated by the speed command value generating means (not shown) provided outside the position sensorless control device of the first embodiment is input to the speed control unit 40 of the microcomputer 22.
The speed control unit 40 generates a γ-axis current command value iγ * and a δ-axis current command value iδ * such that the rotor 12 rotates at a speed according to the analog speed command value ω * a input from the outside, Output to the current control unit 50.

On the other hand, the current sensors 21u and 21v detect currents flowing through the stator windings 11u and 11v, respectively, and create analog u-phase current values iua and v-phase current values iva indicating the current values. The created analog u-phase current value ua and analog v-phase current value iva are input to the current control unit 50 of the microcomputer 22. Current control unit 5
0 indicates that the γ-axis current is the value obtained by superimposing the superimposed wave current value iγa * on the specified γ-axis current value iγ * (iγ * + iγa *), and that the δ-axis current is the value obtained by specifying the δ-axis current value iδ *. Switching signals guh, gul, gvh, gv so that they flow through the stator windings 11u, 11v, 11w.
1, gwh, and gwl are created and output to the drive unit 30.

[Operation of Driving Unit 30] Next, the operation of the driving unit 30 will be described. As shown in FIG.
Controls the voltages of the stator windings 11u, 11v, 11w based on the switching signals guh, gul, gvh, gvl, gwh, gwl input to the pre-drive unit 36. Power supply 31 is upper IGBT 32u, 32v, 32w
And the lower IGBTs 34u, 34v, 34w. The pre-drive unit 36 controls the gate voltage of the upper IGBT 32u to thereby control the switching signal guh.
Is "H" (high level), the upper IGBT 32u is turned on, and when the switching signal guh is "L" (low level), the upper IGBT 32u is turned off. On the other hand, the pre-drive unit 36 controls the gate voltage of the lower IGBT 34u to thereby switch the switching signal gu
Is "H", the lower IGBT 34u is turned on, and when the switching signal gu is "L", the lower IGBT 34u is turned on.
u is turned off. Similarly, the switching signals gvh, gvl, gw
The gate voltages of the upper IGBTs 32v and 32w and the lower IGBTs 34v and 34w are controlled based on h and gwl.

[Operation of the Speed Control Unit 40] Next, the analog speed command value ω * a is input from outside, and the speed control unit 40 generates the γ-axis current command value iγ * and the δ-axis current command value iδ *. Will be described with reference to FIG. The speed control unit 40 is activated at every set time, and the
41, PI control unit 42, and torque / current conversion unit 43
The following operations are performed in this order, and the γ-axis current command value iγ * and the δ-axis current command value iδ * are controlled so that the rotor 12 rotates at a speed according to the externally input analog speed command value ω * a. . The ADC 41 converts the analog speed command value ω * a which is an analog value into the speed command value ω * which is a digital value.
To analog / digital conversion. The PI control unit 42
The torque command value T * is controlled using the proportional-integral control (PI control) so that the estimated speed ωm from the angle estimating unit 60 becomes as indicated by the speed command value ω *. As shown in the following equation (6), the difference between the speed command value ω * and the estimated speed ωm is calculated by the proportional gain KP
The result of the proportional integral control using W and the integral gain KIW is defined as a torque command value T *.

T * = KPW · (ω * −ωm) + KIW · Σ (ω * −ωm) (6)

The torque / current converter 43 is provided with the IPMSM1
A γ-axis current command value iγ * and a δ-axis current command value iδ * are created such that the output torque of 0 becomes the torque command value T *. As shown in the following equation (7), the result of dividing the torque command value T * by a set value KT is used to calculate the current command value amplitude ia *.
And Also, as in the following equation (8), the result of multiplying the current command value amplitude ia * by -sin (β *) is defined as the γ-axis current command value iγ *. On the other hand, as shown in the following equation (9), the result of multiplying the current command value amplitude ia * by cos (β *) is defined as the δ-axis current command value iδ *. Here, β * is the current phase that achieves the maximum output torque or the maximum efficiency when the current command value amplitude ia * is given. Hereinafter, this β * is referred to as a current command value phase β *.

Ia * = T * / KT (7) iγ * = − ia * · sin (β *) (8) iδ * = ia * · cos (β *) (9) )

[Operation of Current Control Unit 50] Next, the operation of the current control unit 50 of the microcomputer 22 will be described with reference to FIG. The current control unit 50 is activated at a certain time period (hereinafter, referred to as a current control period), and the ADC 51
u, 51v, three-phase / two-phase converter 52, voltage command value generator 5
3, the following operations are performed in the order of the two-phase / three-phase converter 54 and the PWM controller 55. The current control unit 50 determines that the γ-axis current is γ
The stator winding 11 has the same configuration as the superimposed wave current command value iγa * superimposed on the shaft current command value iγ * (iγ * + iγa *), and the δ-axis current as the δ-axis current command value iδ *.
u, 11v, 11w
h, gul, gvh, gvl, gwh, gwl are controlled.

The ADC 51u and the ADC 5 of the current control unit 50
1v means that an analog u-phase current value iua and an analog v-phase current value iva are converted into a digital value u-phase current value iu and a v-phase current value iv, respectively.
Perform digital conversion. The three-phase to two-phase converter 52 converts a current value indicating a current flowing through the stator windings 11u, 11v, and 11w into a γ-axis current value iγ on a γ-axis and a δ-axis on a δ-axis which is a rotating coordinate system based on an estimated angle θm. And a current value iδ. In addition, a later-described two-phase / three-phase converter 54 includes a stator winding 11
The inverse of the conversion performed by the three-phase to two-phase converter 52 is performed on the voltages applied to u, 11v, and 11w. In particular,
The three-phase to two-phase converter 52 creates the γ-axis current value iγ and the δ-axis current value iδ as shown in the following equations (10) and (11).

Iγ = {(2)} · {iu · sin (θm + 60 °) + iv · sinθm} (10) iδ = {(2)} · ｛iu · cos (θm + 60 °) + iv · cosθm・ ・ ・ ... (11)

The voltage command value creating section 53 outputs a γ-axis current value iγ
Γ-axis voltage using the proportional integral control (PI control) and the non-interference control so that the superimposed wave current command value iγa * is superimposed on the γ-axis current command value iγ * (iγ * + iγa *). The command value vγ * is controlled. Further, the δ-axis voltage command value v is determined by using the proportional-integral control (PI control) and the non-interference control so that the δ-axis current value iδ becomes the same as the δ-axis current command value iδ *.
Control δ *. The γ-axis voltage command value vγ * is calculated by the following equation (1)
It is calculated by 2). As shown in Expression (12), the difference obtained by subtracting the γ-axis current value iγ from the addition result of the specified γ-axis current value iγ * and the superimposed wave current value iγa * is calculated as the proportional gain KP
D and the integral gain KID perform proportional integral control. The result obtained by multiplying the phase resistance R by the specified γ-axis current value iγ * is added to the result. Further, a result of multiplying the result by the estimated angular velocity ωem, the q-axis inductance Lq, and the specified δ-axis current value iδ * is subtracted to calculate a specified γ-axis voltage value vγ *. Here, the estimated angular velocity ωem is calculated from the estimated velocity ωm.

Vγ * = KPD · {(iγ * + iγa *) − iγ} + KID · {(iγ * + iγa *) − iγ} + R · iγ * −ωem · Lq · iδ * (12)

The specified δ-axis voltage value vδ * is obtained by the following equation (13). As shown in equation (13),
The difference between the specified δ-axis current value iδ * and the δ-axis current value iδ is proportionally integrated controlled by the proportional gain KPQ and the integral gain KIQ. The result obtained by multiplying the phase resistance R by the specified δ-axis current value iδ * is added to the result, and the result obtained by multiplying the estimated angular velocity ωem, the d-axis inductance Ld, and the specified γ-axis current value iγ * is added. Furthermore, the result shows that the estimated angular velocity ωem
And the result of multiplying by the dq-axis winding interlinkage magnetic flux effective value に よ る by the permanent magnet 14 is added to calculate the δ-axis voltage command value vδ *.

Vδ * = KPQ · (iδ * −iδ) + KIQ · Σ (iδ * −iδ) + R · iδ * + ωem · Ld · iγ * + ωem · ψ (13)

The two-phase / three-phase conversion unit 54 provides a specified γ-axis voltage command value vγ * and δ
The δ-axis voltage command value vδ * on the axis is converted into a stationary coordinate system,
A u-phase voltage command value vu *, a v-phase voltage command value vv *, and a w-phase voltage command value vw * to be applied to the stator windings 11u, 11v, 11w are created. The u-phase voltage command value vu *, the v-phase voltage command value vv *, and the w-phase voltage command value vw * are specifically calculated by the following equations (14), (15), and (16).

Vu * = {(2/3)} · {vγ * · cosθ−vδ * · sinθ} (14) vv * = {(2/3)} · ｛vγ * · cos ( θ−120 °) −vδ * · sin (θ−120 °)｝ (15) vw * = {(2/3)} · ｛vγ * · cos (θ + 120 °) −vδ * · sin ( θ + 120 °)｝ (16)

The PWM controller 55 includes a two-phase to three-phase converter 54
Pulse width modulation (PWM: Puls) of the u-phase voltage command value vu *, v-phase voltage command value vv *, and w-phase voltage command value vw * from
e Width Modulation). Specifically, the PWM controller 55 generates a triangular wave having a certain set period and an amplitude of E / 2. Here, the cycle of this triangular wave is the same as the current control cycle, and E is the voltage value of the power supply 31. The triangular wave is compared with the u-phase voltage command value vu *. When the u-phase voltage command value vu * is larger, the switching signal guh is set to “H” and the switching signal gu
To “L”. On the other hand, when the u-phase voltage command value vu * is smaller, the switching signal guh is set to “L” and the switching signal gul is set to “H”. Note that when the states of the switching signals guh and gul transition, a short time during which both the switching signals guh and gul are set to “L” is set (this short time is called a dead time). Similarly, v-phase voltage command value vv * and w-phase voltage command value v
The switching signals gvh, gvl, and gwh, gwl are created based on w *.

[Outline of Operation of Angle Estimating Unit 60] Next, the operation of the angle estimating unit 60 will be described with reference to FIG.
The angle estimating unit 60 corrects the estimated angle so that the result obtained by adding the angle error obtained by the low-speed estimation method and the angle error obtained by the high-speed estimation method at a certain rate converges to zero. By correcting the estimated angle in this manner, the angle estimating unit 60 smoothly switches the angle estimation method between the low speed region and the high speed region. Hereinafter, the angle error (low-speed error) εl by the low-speed estimation method and the angle error (high-speed error) εh by the high-speed estimation method will be described.

[Method of Determining Error εl for Low Speed] First, the method of obtaining the error εl for low speed will be described. Error εl for low speed
Is the estimation method of the above-mentioned conventional example 3 (Japanese Patent Laid-Open No. 10-3230).
No. 99).
That is, the current response on the δ-axis when the superimposed wave current command value iγa * is applied is defined as a low-speed error εl. (A) of FIG.
FIG. 7 is a waveform diagram of a u-phase voltage command value vu * in the first embodiment, and FIG. 7B is a waveform diagram of a u-phase current value iu.
As shown in FIG. 7A, an AC component having a shorter cycle than its fundamental component is superimposed on the phase voltage command value. As a result, as shown in FIG. 7B, a response having the same cycle as the AC component superimposed on the phase voltage command value appears in the phase current.
Hereinafter, the superimposed component and its response are referred to as a superimposed component, a superimposed wave, or a superimposed wave component. Here, the fundamental wave component is a component synchronized with the rotation of the rotor 12,
2 to generate a rotating magnetic field for generating a torque for rotating the rotating magnetic field.

FIG. 8A is a waveform diagram showing the specified γ-axis voltage value vγ * in the first embodiment, and FIG. 8B is a waveform diagram showing the γ-axis current value iγ. (C) and (d) of FIG. 8 are waveform diagrams of the δ-axis current value iδ, respectively. In order to superimpose the AC component shown in FIG. 7A, an AC component having a shorter cycle than the fundamental wave component is superimposed on the specified γ-axis current value iγ *. As a result, as shown in FIG. 8A, an AC component having a shorter cycle than the fundamental component is superimposed on the γ-axis voltage command value vγ * by the proportional operation of the current control. Then, as shown in FIG. 8B, the γ-axis current value iγ has a superimposed component that oscillates in the same cycle as the component superimposed on the specified γ-axis voltage value vγ *. When the angle error Δθ exists (when the angle error Δθ is not 0), as shown in FIG. 8C, the δ-axis current value iδ is the same as the component superimposed on the γ-axis voltage command value vγ *. It has a superimposed component that oscillates periodically. However, when the angle error Δθ is 0, as shown in FIG. 8D, the δ-axis current value iδ does not oscillate and has no superimposed component. Therefore, the response of the δ-axis current is detected at the timing indicated by the dotted line in the waveform diagram shown in FIG. 8C, and the response is defined as a low-speed error εl.

[Method for Obtaining High-Speed Error εh] Next, a method for obtaining the high-speed error εh will be described. The high-speed error εh is disclosed in Japanese Patent Application No. 2000-1763 by the same applicant as the present invention.
No. 9 according to the method described in the specification. That is, the high-speed error εh is obtained by creating a reference value of an induced voltage of a certain phase (induced voltage reference value), calculating an induced voltage value from a phase current value and a phase voltage value, and calculating the calculated induced voltage reference value. It is determined from the difference between the value and the induced voltage value. FIG. 9 shows the first embodiment.
FIG. 6 is a waveform diagram showing an induced voltage value, an induced voltage reference value, and a deviation of a u-phase in FIG. FIG. 9 shows an example in which the induced voltage value is delayed by 20 degrees in electrical angle from the induced voltage reference value. That is, the angle error Δθ (= θ−θm) = − 20 °
It is. In FIG. 9, the amplitude of the induced voltage value is about 9% of the amplitude of the induced voltage reference value (the estimated induced voltage amplitude value em).
An example of 0% is shown.

U-phase induced voltage value (u-phase induced voltage value eu)
And u-phase induced voltage reference value (u-phase induced voltage reference value eum)
Are different from each other, the difference (u-phase deviation εu) is not zero. Therefore, this deviation changes according to the angle error Δθ. Therefore, a high-speed error εh is obtained from the deviation. Here, the phase to be estimated is the estimated angle θm
Selected by. Since the u-phase, v-phase, and w-phase are each shifted by 120 ° in electrical angle, the high-speed error εh is always calculated using the phase in which the influence of the phase difference has the greatest influence on the deviation. That is, the estimated angle θm is 0 ° to 30 electrical degrees.
°, 150 ° to 210 °, and 330 ° to 360 °, the magnitude of the u-phase deviation εu becomes almost maximum,
In the u phase, a high-speed error εh is calculated in this region.
Estimated angle θm is an electrical angle of 90 ° to 150 °, 270 ° to
At 330 °, the magnitude of the v-phase deviation εu becomes almost maximum, so that in the v-phase, the high-speed error εh is calculated in this region. Further, the estimated angle θm is an electrical angle of 30 ° to 90 °.
Since the magnitude of the w-phase deviation εu becomes almost maximum in °, 210 ° to 270 °, in the w-phase, the high-speed error εh is calculated in this region.

As shown in FIG. 9, even when the angle error Δθ is the same, when the estimated angle θm is around 0 °, the u-phase deviation εu
Is positive, and the u-phase deviation εu is negative when the estimated angle θm is around 180 °. Therefore, it is necessary to consider the sign depending on the value of the estimated angle θm. Therefore, the estimated angle θm
When the angle is near 0 °, the value obtained by changing the sign of the u-phase deviation εu, that is, (−εu) is set as the high-speed error. On the other hand, when the estimated angle θm is around 180 °, the u-phase deviation εu is directly used as a high-speed error.

[Principle of Angle Estimation] Next, the principle of angle estimation will be described. In the low-speed range, the estimated angle θm is corrected so that the low-speed error εl converges to zero. On the other hand, in the high-speed range, the estimated angle θm is set so that the high-speed error εh converges to 0.
Is corrected. Further, in the switching range between the low speed range and the high speed range, the estimated angle θm is corrected so that the result of adding the error for low speed εl and the error for high speed εh at a certain ratio converges to 0. Here, when the estimated speed ωm increases, the ratio of the low-speed error εl decreases, and the ratio of the high-speed error εh increases. Thus, the ratio between the low-speed error εl and the high-speed error εh is gradually changed. This ratio (synthesis ratio α) is changed as follows. FIG. 10 is a graph illustrating a relationship between the estimated speed ωm and the combination ratio α in the first embodiment.

As shown in FIG. 10, when the estimated speed ωm is low (less than ωα1), the synthetic ratio α is set to 0, and the estimated angle θm is corrected using only the low-speed error ε1. On the other hand, when the estimated speed ωm is large (when it exceeds ωα4), the combining ratio α is set to 1, and the estimated angle θm is corrected using only the high-speed error εh. In the switching range between the low speed range and the high speed range (the range from ωα1 to ωα4), the synthesis ratio α is gradually changed between 0 and 1 as shown in the graph of FIG. In switching from the low-speed range to the high-speed range, the low-speed error εl and the high-speed error εh are added in the range from ωα3 to ωα4 according to the ratio determined by the combination ratio α, and the sum is converged to 0. The angle θm is corrected. Here, when the estimated speed ωm increases, the synthesis ratio α is increased, the ratio of the low-speed error εl is reduced, and the high-speed error εh
To increase. As described above, in the region from ωα3 to ωα4, the ratio between the low-speed error εl and the high-speed error εh is gradually changed.

On the other hand, when switching from the high-speed region to the low-speed region, in the region from ωα2 to ωα1, the low-speed error ε1 and the high-speed error εh are added at a ratio determined by the synthesis ratio α, and the addition result is 0. The estimated angle θm is corrected so as to converge. Here, when the estimated speed ωm decreases, the synthesis ratio α is reduced, the ratio of the low-speed error εl is increased, and the high-speed error εh is reduced. Thus, from ωα2 to ω
In the area of α1, the error for low speed εl and the error for high speed εh
And gradually change the ratio. As described above, the first embodiment
Has the same relationship as the history loop, as shown in FIG.

FIG. 11 shows the estimated angle θm in the first embodiment.
FIG. 4 is a block diagram illustrating the creation operation. As shown in FIG. 11, the low-speed error εl is multiplied by the low-speed proportional gain κpl and (1−composition ratio α), and the value thereof is set to the high-speed error εh.
Is multiplied by the high-speed proportional gain κph and the combination ratio α to calculate a proportional error εp. The low-speed error εl is multiplied by the low-speed integration gain κil and (1−composition ratio α), and the value is added to a value obtained by multiplying the high-speed error εh by the high-speed integration gain κih and the composition ratio α. hand,
An integration error εi is calculated. The proportional error .epsilon.p and the integral error .epsilon.i calculated as described above are integrated and controlled so as to converge to 0, and the advance amount .theta.d is obtained. The advance amount θd is integrated to create an estimated angle θm.

[Details of Operation of Angle Estimating Unit 60] Next, details of the operation of the angle estimating unit 60 of the microcomputer 22 will be described. The low-speed estimating unit 61 in the angle estimating unit 60 shown in FIG. 5 is activated at each timing shown by a dotted line in the waveform diagram shown in FIG. In addition, the estimation high-speed operation unit 70 of the angle estimation unit 60
It is started up in the same cycle as the current control unit 50, and performs operations to be described later in the order of the combination ratio changing unit 71, the superimposed wave creating unit 72, the high speed estimating unit 73, and the angle / speed creating unit 74. Then, the low-speed error εl obtained by the low-speed estimation unit 61
By correcting the estimated angle θm so that the result of adding the high-speed error εh obtained by the high-speed estimator 73 at a certain ratio converges to 0, the angle estimation method between the low-speed region and the high-speed region can be smoothly performed. Switch. The low-speed estimating unit 61
The operation is performed when the combination ratio α is less than 1, and the low-speed error ε1 is created by using the estimation method in the low-speed range. When the combination ratio α is less than 1, the low-speed estimating unit 61 calculates the δ as shown in the following equation (17) at the timing indicated by the dotted line in FIG.
The difference between the shaft current value iδ and the specified δ-axis current value iδ * is calculated and set as a low-speed error εl. On the other hand, when the combination ratio α is 1, the low-speed error ε1 is unnecessary, and the low-speed estimation unit 61 is not operated at this time.

Εl = iδ−iδ * (17)

The combining ratio changing unit 71 creates a combining ratio α for determining the ratio between the low-speed error εl and the high-speed error εh.
As shown in FIG. 10, when changing from low speed to high speed,
The composition ratio α is created as follows. When the estimated speed ωm is less than a set value ωα3, the combining ratio α is set to 0.
When the estimated speed ωm is larger than a set value ωα4, the combining ratio α is set to 1. When the estimated speed ωm is equal to or more than ωα3 and equal to or less than ωα4, the composition ratio α is obtained by linear interpolation between the coordinates (ωα3, 0) and the coordinates (ωα4, 1).
On the other hand, when changing from a high speed to a low speed, the synthesis ratio α is created as follows. A set value ωα with an estimated speed ωm
When it is less than 1, the combining ratio α is set to 0. When the estimated speed ωm is larger than a set value ωα2, the combining ratio α is set to 1. When the estimated speed ωm is equal to or more than ωα1 and equal to or less than ωα2,
The combination ratio α is represented by coordinates (ωα1, 0) and coordinates (ωα2, 1)
And is obtained by linear interpolation.

Superimposed wave creating section 72 in estimation high-speed operation section
When the motor speed is slow and the low-speed error ε1 is required, the superimposed wave current command value iγ is used for the estimation method in the low-speed range.
Create a *. On the other hand, when the speed of the motor is high and the low-speed error εl is unnecessary, the superimposed wave current command value iγa * is set to 0. As shown in the block diagram of FIG.
If it is less than the predetermined value, the estimated speed ωm is calculated using both the low-speed error ε1 and the high-speed error εh. At this time, as shown in the following equation (18), the superimposed wave current command value iγa * is an AC component having a shorter cycle than the fundamental wave component. In Expression (18), Aiγa is the amplitude of the superimposed wave current command value iγa *, ωea is the angular velocity of the superimposed wave current command value, and t is time. On the other hand, as can be understood from FIG. 11, when the combination ratio α is 1, the low-speed error εl becomes unnecessary. Therefore, there is no need to superimpose an AC component having a shorter cycle than the fundamental component on the specified γ-axis current value iγ *. Therefore, the specified γ-axis current value iγ * is set to 0 as in the following equation (19).

Iγa * = Aiγa · sin (ωea · t) (when α <1) (18) iγa * = 0 (19)

High-speed estimation section 73 of estimation high-speed operation section 70
Creates the high-speed error εh using the estimation method in the high-speed range. First, as shown in the following equation (20), a w-phase current value iw is obtained from a u-phase current value iu and a v-phase current value iv. Next, as shown in the following equations (21), (22), and (23), the induced voltages of the u-phase, v-phase, and w-phase are calculated,
The u-phase induced voltage eu, the v-phase induced voltage ev, and the w-phase induced voltage ew are calculated. Here, d / dt represents a time derivative, and dθ / dt appearing in the calculation of the derivative related to the trigonometric function is obtained by converting the estimated speed ωm into an electrical angular speed.
Also, d (iu) / dt, d (iv) / dt, d (i
w) / dt is obtained by first-order Euler approximation. Further, R
Is the resistance per stator winding phase, la is the leakage inductance per stator winding phase, La is the average effective inductance per stator winding phase, and La
s is the amplitude of the effective inductance per stator winding phase.

Iw = − (iu + iv) (20) eu = vu * −R · iu− (la + La) · d (iu) / dt−Las · cos (2θm) · d (iu) / dt −Las · iu · d {cos (2θm)} / dt + 0.5 · La · d (iv) / dt − Las · cos (2θm−120 °) · d (iv) / dt − Las · iv · d {Cos (2θm−120 °)} / dt + 0.5 · La · d (iw) / dt−Las · cos (2θm + 120 °) · d (iw) / dt−Las · iw · d ｛cos (2θm + 120 °) )｝ / Dt (21)

Ev = vv * −R · iv− (la + La) · d (iv) / dt−Las · cos (2θm + 120 °) · d (iv) / dt−Las · iv · d {cos (2θm + 120 °)} /Dt+0.5·La·d(iw)/dt−Las·cos(2θm)·d(iw)/dt−Las·iw·d{cos(2θm)}/dt+0.5·La· d (iu) / dt−Las · cos (2θm−120 °) · d (iu) / dt−Las · iu · d {cos (2θm−120 °)} / dt (22)

Ew = vw * −R · iw− (la + La) · d (iw) / dt−Las · cos (2θm−120 °) · d (iw) / dt−Las · iw · d ｛cos (2θm− 120 °)｝ / dt + 0.5 · La · d (iu) / dt−Las · cos (2θm + 120 °) · d (iu) / dt−Las · iu · d {cos (2θm + 120 °)} / dt + 0.5 · La · d (iv) / dt−Las · cos (2θm) · d (iv) / dt−Las · iv · d {cos (2θm)} / dt (23)

Next, the phase having the largest deviation is used as the phase used for estimation (estimated phase). As shown in FIG.
When the estimated angle θm is 0 ° or more and less than 30 °, the estimated phase index η is set to 0. When the estimated angle θm is 30 ° or more and less than 90 °, the estimated phase index η is set to 1. Estimated angle θm is 90 °
If it is less than 150 °, the estimated phase index η is set to 2. Hereinafter, every time the estimated angle θm changes by 60 °, the estimated phase index η is increased by one. Then, the estimated angle θm is 27
When it is 0 ° or more and less than 330 °, the estimated phase index η is set to 5. Then, when the estimated angle θm is not less than 330 ° and less than 360 °, the estimated phase index η is set to 0. Here, when the estimated phase index η = 0, 3, the estimated phase is the u phase, and the estimated phase index η =
In the case of 1, 4, the estimated phase is the w phase, and the estimated phase index η = 2,
At 5, the estimated phase is the v phase.

Next, a high-speed error εh is obtained. As shown in the following equation (24), based on the estimated phase index η, the induced voltage value of each phase (u-phase induced voltage value eu, v-phase induced voltage value e
v, w phase induced voltage value ew) and the induced voltage reference value (u
Phase induced voltage reference value eum, v phase induced voltage reference value evm,
From the difference from the w-phase induced voltage reference value ewm), the error for high speed ε
Find h. Here, even for the same angle error Δθ, the sign of the difference between the induced voltage value and the induced voltage reference value is different, so the sign is considered in the estimated angle θm. In addition, the induced voltage reference values (u-phase induced voltage reference value eum, v-phase induced voltage reference value evm, w-phase induced voltage reference value ewm) of each phase are represented by the following equation (25). In equation (25), em
Is the induced voltage reference value of each phase (u-phase induced voltage reference value eum,
V-phase induced voltage reference value evm, w-phase induced voltage reference value ew
m), for example, the induced voltage value of each phase (u-phase induced voltage value eu, v-phase induced voltage value ev, w-phase induced voltage value e
It is assumed that a low-pass filter is applied to the square root of the sum of squares of w).

Εh = − (eu−eum) (when η = 0) εh = (ew−ewm) (when η = 1) εh = − (ev−evm) (when η = 2) εh = ( eu−eum) (when η = 3) εh = − (ew−ewm) (when η = 4) εh = (ev−evm) (when η = 5) (24) eum = −em・ Sin (θm + α) evm = −em ・ sin (θm + α−120 °) ewm = −em ・ sin (θm + α−240 °) (25)

Angle / velocity creating section 7 of estimated high-speed operation section 70
4 creates an estimated angle θm and an estimated speed ωm. First, the angle / speed creating unit 74 corrects the estimated angle θm so that the result of adding the low-speed error εl and the high-speed error εh at a certain rate converges to 0. As shown in the block diagram of FIG. 11 and the following equation (26), the low-speed error εl is multiplied by the low-speed proportional gain κpl and (1−composition ratio α),
The value is added to a value obtained by multiplying the high-speed error εh by the high-speed proportional gain κph and the combination ratio α to calculate the proportional error εp. The low-speed error εl is multiplied by the low-speed integral gain κil and (1−composition ratio α), and the value is added to the value obtained by multiplying the high-speed error εh by the high-speed integral gain κih and the composition ratio α. Thus, the integration error εi is calculated. Then, as shown in FIG. 11 and the following equation (28), the integration error εi is integrated and added to the proportional error εp, and the advance amount θd
Is obtained. Further, as shown in FIG. 11 and the following equation (29), the advance amount θd is integrated, and the estimated angle θm is created.

Εp = εl · κpl · (1-α) + εh · κph · α (26) εi = εl · κil · (1-α) + εh · κih · α (27) θd = Εp + Σεi (28) θm = Σθd (29)

The angle / speed generating unit 74 generates an estimated speed ωm by passing the calculated advance amount θd through a primary digital low-pass filter (LPF). Specifically, the angle / speed creating unit 74 calculates the following equation (3
0) is performed to generate an estimated speed ωm.
In Expression (30), ωm (n) is the current estimated speed, and ωm (n−1) is the previous estimated speed. Also, K
TPW is a coefficient for converting the advance amount θd into a unit of the estimated speed ωm. Further, KLW is a coefficient of a low-pass filter, takes a value from 0 to 1, and the smaller the value, the greater the effect of the low-pass filter.

Ωm (n) = KLW · (KTPW · θd) + (1−KLW) · ωm (n−1) (30)

[Effects of Position Sensorless Control Device of First Embodiment] Next, effects of the position sensorless control device of the first embodiment will be described. The estimated angle θm may deviate from the actual angle θ due to parameter error, current sensor error, influence of dead time, difference between current command value and actual current, difference between voltage command value and actual voltage, and calculation delay. is there. This shift differs depending on the estimation method. for that reason,
The estimated angle θm estimated only from the low-speed region estimation method may be different from the estimated angle θm estimated only from the high-speed region estimation method. As described above, when the two are greatly different depending on the estimation method, if the estimation method in the low-speed region and the estimation method in the high-speed region are instantaneously switched, the deviation of the estimation angle θm suddenly becomes large and the step-out may occur. In the position sensorless control device of the first embodiment, the low-speed error ε is
1 and the high-speed estimator 73 generates the high-speed error εh. In the first embodiment, the estimated speed ω
A composition ratio α that varies with m is created. And
The estimated angle θm is corrected by using the proportional integral control so that the result of adding the error for low speed εl and the error for high speed εh at a certain rate converges to 0 based on the created combination ratio α.

As described above, according to the first embodiment, the estimated angle θm is corrected so that the result of adding the error for low speed εl and the error for high speed εh at a certain rate converges to 0. Thus, it is possible to obtain a position sensorless control device for a synchronous motor capable of smoothly switching the angle estimation method with respect to the range.

Embodiment 2 Next, Embodiment 2 according to the present invention.
The synchronous motor position sensorless control device will be described with reference to the accompanying drawings. The position sensorless control device for a synchronous motor according to the first embodiment described above uses the calculated estimated speed ωm
The position sensorless control device according to the second embodiment is configured to change the combined ratio α by calculating each time.
In this case, the calculation processing is reduced to shorten the calculation time. The position sensorless control device for a synchronous motor according to the second embodiment creates a change amount αd for each operation cycle of the low-speed estimation method, and changes the composite ratio α for each operation cycle of the high-speed estimation method using the change amount αd. By configuring to
This is to reduce the calculation time.

[Configuration of Position Sensorless Control Device of Second Embodiment] First, the configuration of the position sensorless control device of the synchronous motor of the second embodiment will be described. FIG. 12 is a block diagram illustrating a configuration of the position sensorless control device according to the second embodiment.
The position sensorless control device according to the second embodiment is different from the position sensorless control device according to the first embodiment in that
The angle estimation unit 260 in the microcomputer 222. Other configurations in the second embodiment are the same as those in the first embodiment, and thus description thereof will be omitted. FIG. 13 is a block diagram illustrating a configuration of the angle estimation unit 260 of the microcomputer 222 according to the second embodiment. Of the configuration of the angle estimating unit 260 according to the second embodiment,
The estimated high-speed operation unit 270 is different from the estimated high-speed operation unit 70 (FIG. 5) of the first embodiment. Further, the angle estimating section 260 of the second embodiment is provided with a change amount creating section 262. Then, in the estimation high-speed operation unit 270, the combining ratio changing unit 27
1 has a different configuration from the combining ratio changing unit 71 of the first embodiment. Other configurations are the same as those of the first embodiment, and the same reference numerals are given and the description thereof is omitted.

[Operation of Position Sensorless Control Device of Second Embodiment] Next, the operation of the position sensorless control device of the synchronous motor of the second embodiment will be described. The change amount creation unit 262 of the angle estimation unit 260 calculates the estimated speed ωm from the speed creation unit 75.
Is input, the change amount αd is created, and the synthesis ratio changing unit 271 is input.
Output to Further, the synthesis ratio changing unit 271 calculates the change amount αd
Is input to generate the combined ratio α and output it to the superimposed wave generating unit 72. The input / output operations of the components other than the change amount creating unit 262 and the composition ratio changing unit 271 are the same as those of the first embodiment, and the description thereof will be omitted. First, a method of creating the combination ratio α in the second embodiment will be described. In the first embodiment described above, the estimation high-speed operation unit 70 is activated, and the synthesis ratio changing unit 7
1 is executed, the composite ratio α is calculated based on the estimated angle ωm.
Had been created. In the second embodiment, the change amount creation unit 262 started at a longer interval than the start of the estimated high-speed operation unit 270 is configured to create the change amount αd. Then, every time the estimated high-speed operation unit 270 is activated, the combination ratio changing unit 271 changes the combination ratio α by the change amount αd. Thus, the calculation time required for the estimated high-speed operation unit 270 is reduced.

FIG. 14 is an explanatory diagram showing a method for creating the combination ratio α in the second embodiment. In the graph shown in FIG. 14, the activation period of the change amount
This is eight times as long as the 70 startup cycle. Α0 (hereinafter referred to as a synthesis ratio reference value α0) as a reference for creating the change amount αd is formed every time the change amount creation unit 262 is started. FIG.
In 4, the composite ratio reference value α0 at the time of startup is represented by a white point (○). In FIG. 14, the combination ratio α at the time of activation of the estimated high-speed operation unit 270 is represented by a black point (●). The synthesis ratio reference value α0 is calculated from the estimated angle ωm, as shown in FIG. α0 (i) is the synthesis ratio reference value when the change amount creation unit 262 is started this time. Also, α0 (i
-1) is the combination ratio reference value at the time when the change amount creation unit 262 was last activated. Further, α0 (i + 1) is a value that is expected to be the synthesis ratio reference value when the change amount creation unit 262 is started next time, and is extrapolated from α0 (i−1) and α0 (i). As shown in FIG. 14, the estimated high-speed operation unit 270 operates eight times during a period in which the combination ratio reference value is from α0 (i) to α0 (i + 1). Therefore, as shown in the following equation (31), the change amount creation unit 262 calculates α0 (i−1) to α0 (i−1).
The value obtained by dividing the deviation to 0 (i) by 8 is used as the change amount αd. Then, as indicated by a black dot (●) in FIG. 14, the combination ratio changing unit 271 changes the combination ratio α by this change amount αd.

Αd = {α0 (i) −α0 (i−1)} / 8 (31)

That is, the change amount creating section 262 and the composition ratio changing section 271 are operated as follows. The change amount creation unit 262 is activated at a cycle eight times the activation cycle of the estimated high-speed operation unit 270. First, as shown in FIG. 10, a combined ratio reference value α0 is created from the estimated angle ωm. According to the relationship between the estimated angle ωm and the combination ratio α in FIG. 10, the first embodiment is used to create the combination ratio α in the above-described first embodiment, but is used to create the combination ratio reference value α0 in the second embodiment. From α0 (i) and α0 (i−1) created in this way, equation (31) is obtained.
Is used to generate the change amount αd. The combination ratio changing unit 271 of the second embodiment is started at the same timing as the combination ratio changing unit 71 of the first embodiment. As shown in the following equation (32), the current synthetic ratio α is obtained by changing the previous synthetic ratio α by the change amount αd
Only change. In Expression (32), α (i) is the composition ratio created this time by the composition ratio changing unit 271, and α (i)
-1) is the combination ratio previously created by the combination ratio changing unit 271.

Α (i) = α (i−1) + αd (32)

[Effects of the Position Sensorless Control Device of the Second Embodiment] Next, the effects of the position sensorless control device of the synchronous motor according to the second embodiment configured and operated as described above will be described. The position sensorless control device of the second embodiment has substantially the same configuration as that of the first embodiment, and thus has the same effect as that of the first embodiment. That is, the position sensorless control device according to the second embodiment has an effect that the angle estimation method between the low speed range and the high speed range can be smoothly switched.

In the position sensorless control device according to the first embodiment, the estimation high-speed operation unit 70 is activated, and the combined ratio changing unit 7
Each time 1 is executed, the synthesis ratio α is created from the estimated angle ωm. In the position sensorless control device according to the second embodiment, the change amount creation unit 262 started at a longer interval than the estimated high-speed operation unit 270 creates the change amount αd. Then, every time the estimated high-speed operation unit 270 is activated, the synthesis ratio changing unit 271 changes the synthesis ratio α by the change amount αd, so that the calculation time required for the estimated high-speed operation unit 270 is greatly reduced. As described above, according to the second embodiment, a change amount αd is created, and by changing the combination ratio α by the change amount αd, a position sensorless control device for a synchronous motor with a short operation time can be realized.

In the first and second embodiments,
As the low-speed estimation method, the estimation method described in the above-mentioned conventional example 3 (the estimation method disclosed in JP-A-10-323099) is used. Although the system is described in the specification of Japanese Patent Application Laid-Open No. 2000-17639 and the high-speed estimation system described in the first embodiment is described as an example, the present invention is not limited to these estimation systems, and other general methods are used. The estimation method used in general may be used. The essence of the present invention is that the error is obtained from the low-speed estimation method and the high-speed estimation method, and the estimated angle θm is adjusted so that the result of adding the low-speed error εl and the high-speed error εh at a certain ratio converges to 0. It is to correct. Therefore, even if another estimation method is used, it is included in the present invention.

The low-speed estimation method and the high-speed estimation method used in the first and second embodiments are methods for correcting the estimated angle so that the angle error converges to 0 when used alone. The present invention reduces the angle error to zero.
However, the present invention can also be applied to a method that does not correct the estimated angle so as to converge. For example, in the low-speed estimation method,
Consider the case where the estimated angle is corrected so that the angle error converges to 0 as in (1) and the estimated angle is directly created as in the first embodiment in the high-speed estimation method. In the low-speed estimation method, a low-speed error εl is created as in the first embodiment. On the other hand, the high-speed estimation method creates an estimated angle (high-speed estimated angle).
Further, a difference between the actually used estimated angle (the estimated angle used in the current control unit 50) and the high-speed estimated angle is defined as a high-speed error εh. Then, as in the first embodiment, the result of adding the error for low speed εl and the error for high speed εh at a certain ratio is 0.
The estimated angle θm is corrected so as to converge. Similarly, the present invention can be applied to a case where the low-speed estimation method directly obtains the estimated angle and the high-speed estimation method corrects the estimated angle so that the angle error converges to zero.
Further, the present invention can be applied even when the low-speed estimation method directly obtains the estimated angle and the high-speed estimation method directly obtains the estimated angle.

In the second embodiment, the example has been described in which the activation cycle of the change amount creation unit 262 is eight times the activation cycle of the estimated high-speed operation unit 270, but the present invention is not limited to this value. Further, in the second embodiment, the example has been described in which the change amount αd is created from the combination ratio reference value α0, but the present invention is not limited to this. The second embodiment is characterized in that a change amount αd is created and the combination ratio α is changed based on the change amount αd, and this feature can be variously modified. For example, even if the change amount αd is constant, the synthesis ratio α gradually changes, so that it is possible to smoothly switch the angle estimation method between the low-speed region and the high-speed region. In the first and second embodiments, an example in which the low-speed error εl and the high-speed error εh are added at a certain ratio and multiplied by a proportional gain and an integral gain, respectively, has been described separately. The invention is not limited to such arithmetic processing. For example, as shown in the following equation (33), the error ε1 for low speed and the error εh for high speed are added at a certain ratio to create an error ε after addition. Next, the following equation (3)
As shown in 4), even if the post-addition error ε converges to 0, the post-addition error ε is proportionally integrated controlled by the proportional gain κp and the integral gain κi to produce the advance amount θd. Good.

Ε = (1−α) · εl + α · εh (33) θd = κp · ε + Σκi · ε (34)

In the first and second embodiments, the IP
Although the example in which the MSM is controlled has been described, the synchronous motor of the present invention is not limited to the IPMSM. The present invention relates to, for example, SPMSM (Surface Permanent Magnet Sync).
hronous Motor: A configuration for controlling a surface magnet type synchronous motor may be used, and a SynRM (Synchronous Relucta
nce Motor (synchronous reluctance motor) may be controlled.

[0091]

As is apparent from the detailed description of the embodiments, the present invention has the following effects.
According to the present invention, it is possible to smoothly switch the angle estimation method between the low-speed region and the high-speed region by correcting the estimated angle so that the result of adding the error for the low speed and the error for the high speed at a certain ratio converges to 0. And a control device and control device for a synchronous motor that can perform the position sensorless operation. Also,
According to the present invention, it is possible to obtain a position sensorless control method and a control device for a synchronous motor with a short operation time by creating a change amount of the synthesis ratio and changing the synthesis ratio by the change amount.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a position sensorless control device for a synchronous motor according to a first embodiment of the present invention.

FIG. 2 is a circuit diagram illustrating a configuration of a driving unit according to the first embodiment.

FIG. 3 is a block diagram illustrating a configuration of a speed control unit according to the first embodiment.

FIG. 4 is a block diagram illustrating a configuration of a current control unit according to the first embodiment.

FIG. 5 is a block diagram illustrating a configuration of an angle estimating unit according to the first embodiment.

FIG. 6 is an explanatory diagram of a coordinate system according to the first embodiment.

FIG. 7 is a waveform diagram of a u-phase voltage command value (a) and a u-phase current value (b) in the first embodiment.

FIG. 8 is a waveform chart of a γ voltage command value (a), a γ axis current value (b), and a δ axis current value (c) and (d) in the first embodiment.

FIG. 9 is a waveform diagram showing a u-phase induced voltage value, an induced voltage reference value, and a deviation in the first embodiment.

FIG. 10 is a graph showing a relationship between a synthetic ratio and an estimated speed in the first embodiment.

FIG. 11 is a block diagram illustrating an operation of creating an estimated angle in the first embodiment.

FIG. 12 is a block diagram illustrating a configuration of a position sensorless control device for a synchronous motor according to a second embodiment.

FIG. 13 is a block diagram illustrating a configuration of an angle estimating unit according to a second embodiment.

FIG. 14 is an explanatory diagram illustrating a method for creating a combination ratio in the second embodiment.

[Explanation of symbols]

 Reference Signs List 10 IPMSM 21u, 21v Current sensor 22, 222 Microcomputer 30 Drive unit 40 Speed control unit 50 Current control unit 60, 260 Angle estimation unit 61 Low-speed estimation unit 70, 270 Estimation high-speed operation unit 71, 271 Combination ratio change unit 73 High-speed Estimation unit 74 Angle / speed creation unit 262 Change amount creation unit

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Toru Tazawa 1006 Kadoma Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. 72) Inventor Yoshiteru Ito 1006 Kazuma Kadoma, Kadoma City, Osaka Prefecture F-term (reference) 5H560 BB04 BB07 BB17 DA12 DB12 DC01 EB01 RR10 SS01 TT01 TT08 TT11 TT15 TT18 UA06 XA02 XA04 DD1212 EE11 EE19 FF07 FF08 GG02 GG04 HA04 HB02 JJ03 JJ04 JJ09 JJ16 JJ17 JJ18 JJ24 JJ26 LL14 LL22 LL25 LL41

Claims (10)

[Claims] 1. A position sensorless control method for estimating a rotor angle in a synchronous motor to form an estimated angle, and controlling the synchronous motor based on the estimated angle, the method being formed using a first estimation method. Periodically creating a first error that varies according to the estimation error of the estimated angle, a second error that varies according to the estimation error of the estimated angle formed using a second estimation method. Periodically generating the first error and the second error at a predetermined ratio to generate a post-addition error, and such that the post-addition error converges to zero. Correcting the estimated angle. 2. The position of the synchronous motor according to claim 1, wherein a combination ratio, which is an addition ratio of the first error and the second error, is changed according to a rotation speed of the rotor. Sensorless control method. 3. The method according to claim 1, further comprising: creating a temporal change amount of the composition ratio; and changing the composition ratio based on the variation amount in synchronization with a cycle of creating the second error. 3. The method according to claim 2, wherein the synchronous motor has a position sensorless control. 4. The method according to claim 1, further comprising: creating an estimated speed of the rotor based on the formed estimated angle; and creating an acceleration of the rotor based on the estimated speed. 4. The method according to claim 3, wherein the position sensorless control is performed. 5. The system according to claim 2, further comprising a hysteresis loop in a relationship between the rotational speed of the rotor and the composite ratio, wherein the composite ratio is changed according to the rotational speed.
A method for controlling a position sensorless motor of a synchronous motor according to claim 3. 6. A position sensorless control device for a synchronous motor, comprising: an estimated angle creating unit that creates an estimated angle of a rotor in a synchronous motor; and a driving unit that drives the synchronous motor based on the estimated angle. A first error creating unit that periodically creates a first error that changes in accordance with an estimation error of the estimated angle formed by using a first estimation method; A second error creating unit that periodically creates a second error that changes in accordance with an estimation error of the estimated angle formed using an estimation method; and the first error in accordance with a rotation speed of the rotor. And the second
A post-addition error generating means for generating a post-addition error by substantially adding the first error and the second error by changing a synthesis ratio which is an addition ratio with the post-addition error; A position sensorless control device for a synchronous motor, comprising: an estimated angle correction means for correcting the estimated angle so that the error converges to zero. 7. The post-addition error creating means has a hysteresis loop in a relationship between the rotation speed of the rotor and the synthesis ratio, and is configured to change the synthesis ratio in accordance with the rotation speed. 7. The position sensorless motor control device for a synchronous motor according to claim 6, wherein: 8. The synthesizing ratio is changed based on the change amount in synchronization with an operation cycle of the change amount generating means for generating a temporal change amount of the synthesizing ratio and an operation cycle of the second error generating means. 7. The control device according to claim 6, further comprising a combination ratio changing unit. 9. The change amount creating means creates an estimated speed of the rotor based on the formed estimated angle, creates an acceleration of the rotor based on the estimated speed, and creates the change amount of the rotor based on the acceleration. The position sensorless control device for a synchronous motor according to claim 8, wherein the control device is configured to generate the following. 10. The method according to claim 1, wherein the change amount creating means has a hysteresis loop in a relationship between the rotation speed of the rotor and the change amount, and is configured to change the change amount according to the rotation speed. The position sensorless motor control device for a synchronous motor according to claim 8, wherein