JP4691915B2 - Motor position estimation method, motor position estimation apparatus, inverter control method, and inverter control apparatus - Google Patents

Description

  The present invention relates to a technique for estimating the position of a motor, and is applied, for example, when estimating the rotational position of an embedded permanent magnet type brushless DC motor.

  Conventionally, as a technology for estimating the position of the rotor of the motor (hereinafter simply referred to as “position”) without using a sensor, the high-frequency signal is input to the motor using the fact that the inductance of the fixed coordinate system varies depending on the position. Then, a method based on the high-frequency component of the current output from the motor (for example, see Non-Patent Documents 1 and 2 below) and a method based on the back electromotive voltage generated by the motor (for example, see Non-Patent Document 3 below) are proposed. Has been.

  The former method does not require a back electromotive force. Therefore, the position can be estimated even in a situation where the motor rotates at low speed or stops rotating. However, if the rotational speed increases, the frequency of the necessary high-frequency component increases.

  On the other hand, a technique for switching a method for estimating the position of the motor according to the number of rotations of the motor has been proposed (for example, see Patent Documents 1 to 3 below).

JP 2002-315386 A Japanese Patent No. 3486326 Japanese Patent Laid-Open No. 7-177788 Kazunori Yamada and two others, “Position sensorless speed control method of salient pole type PM motor including low speed region”, IEEJ Semiconductor Power Conversion Study Group Material, January 1997, SPC-97-13, p75-82 Takehiro Kimihiro and two others, “Robust rotor position / speed sensorless control method of PM rotor focusing on harmonic instantaneous reactive power”, 1999 IEEJ National Conference on Industrial Applications, NoII-188, 1999, p. 47-50 Ken Zhen Chen and three others, "Sensorless position estimation method and stability of salient pole type brushless DC motor", 1998 IEEJ National Conference on Industrial Applications, Vol. 1, No. 1 59, 1998, p. 179-182

  However, there is a problem that if the two methods described above are simply switched, the estimated position value will fluctuate rapidly.

  Accordingly, an object of the present invention is to provide a technique that does not cause a sudden change in an estimated position value when switching between a position estimation technique suitable for low-speed rotation and a position estimation technique suitable for high-speed rotation.

A first aspect of the motor position estimation method according to the present invention is a motor position estimation method that employs the first technique (11) and the second technique (12) for estimating the position of the motor (30). is there. The estimation of the position by the second method is influenced by a robust parameter in the estimation of the position in the first method. The position estimated by the first method is set as a first position candidate (θ ^ _LOW ), and the position estimated by the second method is set as a second position candidate (θ ^ _HI ). (A) adopting the first position candidate as an estimated value (θ ^ _re ) of the position when the rotational speed (ω ^ _re ) of the motor is equal to or less than a predetermined threshold (C) (S107); (B) When the rotational speed shifts from a value equal to or less than a predetermined threshold (C) to a value exceeding the threshold, the second position candidate is equal to the first position candidate. Changing the parameter (Lq ^) in the rotating coordinate system (S105, S106), and adopting the second position candidate obtained thereafter as the estimated value (S108).

  Preferably, in the first method (11), the position is estimated based on the fact that the inductance in the fixed coordinate system of the motor (30) varies depending on the position. In the second method (12), the position is estimated based on a counter electromotive voltage generated by the motor.

A second aspect of the motor position estimation method according to the present invention is the motor position estimation method according to the first aspect, wherein the parameter is an estimated value (Lq ^) of the q-axis inductance of the motor. In the step (b), when the second position candidate (θ ^ _HI ) is larger than the first position candidate (θ ^ _LOW ), the estimated value of the q-axis inductance of the motor is increased ( S105), when the second position candidate is less than or equal to the first position candidate, a step of reducing the estimated value of the q-axis inductance (S106), and the second position candidate being the first position candidate Step (S107) is employed in which the first position candidate is employed as the estimated position value (θ ^ _re ) until it becomes equal to the candidate.

A third aspect of the motor position estimation method according to the present invention is a motor position estimation method that employs the first technique (11) and the second technique (12) for estimating the position of the motor (30). is there. The position estimated by the first method is set as a first position candidate (θ ^ _LOW ). The position estimated by the second method is set as a second position candidate (θ ^ _HI ). (A) In a situation where the rotational speed (ω ^ _re ) of the motor never exceeds the first threshold (C) from the time of startup, the first position candidate is adopted as the estimated value (θ ^ _re ) of the position Step (S210), and (b) when the rotational speed shifts from a value equal to or lower than the first threshold value to a value exceeding the first threshold value (b-1) the second position. A step (S205) of adopting, as the estimated value, a value that gradually shifts from the first position candidate to the candidate; and (b-2) after the second position candidate is adopted as the estimated value. Comprises the step of continuously adopting the second position candidate as the estimated value (S204). At a rotational speed higher than the first threshold, the second method has higher accuracy of position estimation than the first method.
In the step (b-1), the estimated value is a value obtained by linear interpolation between the first position candidate and the second position candidate as time passes, and the first position candidate A linear transition is made from (θ ^ _LOW ) to the second position candidate (θ ^ _HI ).

  Preferably, in the first method (11), the position is estimated based on the fact that the inductance in the fixed coordinate system of the motor (30) varies depending on the position. In the second method (12), the position is estimated based on a counter electromotive voltage generated by the motor.

A fourth aspect of the motor position estimation method according to the present invention is a motor position estimation method that employs the first method (11) and the second method (12) for estimating the position of the motor (30). is there. The position estimated by the first method is set as a first position candidate (θ ^ _LOW ). The position estimated by the second method is set as a second position candidate (θ ^ _HI ). (A) In a situation where the rotational speed (ω ^ _re ) of the motor never exceeds the first threshold (C) from the time of startup, the first position candidate is adopted as the estimated value (θ ^ _re ) of the position Step (S210), and (b) when the rotational speed shifts from a value equal to or lower than the first threshold value to a value exceeding the first threshold value (b-1) the second position. A step (S205) of adopting, as the estimated value, a value that gradually shifts from the first position candidate to the candidate; and (b-2) after the second position candidate is adopted as the estimated value. Comprises the step of continuously adopting the second position candidate as the estimated value (S204). At a rotational speed higher than the first threshold, the second method has higher accuracy of position estimation than the first method. And, in the case where the operation proceeds to (c) the rotational speed (ω ^ _re) from a value higher than the second threshold value (C), the second threshold value the following values, (c-1) wherein the Adopting a value that gradually shifts from the second position candidate (θ ^ _HI ) to the first position candidate (θ ^ _LOW ) as the estimated value (S211), (c-2) After the position candidate of 1 is adopted as the estimated value, the method further includes a step (S210) of continuing to adopt the first position candidate as the estimated value. At a rotational speed lower than the second threshold, the first method has a higher estimation accuracy of the position than the second method.
In the step (c-1), the estimated value is a value obtained by linear interpolation between the first position candidate and the second position candidate as time elapses, and the second position candidate ( It changes linearly from θ ^ _HI ) to the first position candidate (θ ^ _LOW ).

A fifth aspect of the motor position estimation method according to the present invention is a motor position estimation method employing the first method (11) and the second method (12) for estimating the position of the motor (30). is there. The position estimated by the first method is set as a first position candidate (θ ^ _LOW ). The position estimated by the second method is set as a second position candidate (θ ^ _HI ). (A) In a situation where the rotational speed (ω ^ _re ) of the motor never exceeds the first threshold (C) from the time of startup, the first position candidate (θ ^ _LOW ) is set to the first cutoff frequency ( ( ¹ ) Step (S308, S312) which is adopted as an estimated value (θ ^ _re ⁺ ) of the position by performing a low-pass transmission filtering process in f1), and (b) When shifting to a value exceeding the first threshold, (b-1) In the predetermined period (S304), the second position candidate (θ ^ _HI ) is lower than the first cutoff frequency. A step (S303, S306) of performing low-pass transmission filtering processing at the second cutoff frequency (f2) and adopting the estimated value of the position; and (b-2) after the predetermined period has elapsed, A candidate position is a third higher than the second cutoff frequency. Cutoff at a frequency (f3) was treated low-pass filter and a step of employing (S303, S305) as an estimate of the position. At a rotational speed higher than the first threshold, the second method has higher accuracy of position estimation than the first method.

  Preferably, in the first method (11), the position is estimated based on the fact that the inductance in the fixed coordinate system of the motor (30) varies depending on the position. In the second method (12), the position is estimated based on a counter electromotive voltage generated by the motor.

A sixth aspect of the motor position estimation method according to the present invention is the motor position estimation method according to the fifth aspect, wherein (c) the rotation speed of the motor is higher than a second threshold value. When the value shifts to a value equal to or lower than the second threshold value, (c-1) the first position candidate (θ ^ _LOW ) is determined from the third cutoff frequency (f3) in a predetermined period. Adopting a low-pass transmission filtering process at a lower fourth cut-off frequency (f4) and adopting (S308, S311) as the estimated position value (θ ^ _re ), (c-2) elapse of the predetermined period Later, the method further includes a step of subjecting the first position candidate to low pass transmission filtering processing at the first cutoff frequency (f1) and adopting it as an estimated value of the position (S308, S312). At a rotational speed lower than the second threshold, the first method has a higher estimation accuracy of the position than the second method.

A seventh aspect of the motor position estimation method according to the present invention is the motor position estimation method according to the fourth and sixth aspects, wherein the first threshold value is larger than the second threshold value.

An eighth aspect of the motor position estimation method according to the present invention is the motor position estimation method according to the first to seventh aspects, in the first method, more than the rotational speed (ω _re ). A high frequency signal (v ^* _h ) having a high frequency is input to the motor, and the position of the motor is estimated based on a high frequency component of a current flowing through the motor.

The motor position estimation method according to any of the first to eighth aspects can be employed as an inverter control method.

The first mode of the motor position estimation device (1) according to the present invention estimates the position of the motor (30) and outputs it as a first position candidate (θ ^ _LOW ) (11 ), A second position estimation unit (12) that estimates the position and outputs it as a second position candidate (θ ^ _HI ), a rotational speed (ω ^ _re ) of the motor, and a predetermined threshold (C) And a switching determination unit (14) that performs a comparison between the first position candidate and the second position candidate based on the comparison result of the switching determination unit. switching means (13) for outputting as θ ^ _re ). The estimation of the position in the second position estimation unit is influenced by a robust parameter (Lq ^) in the estimation of the position in the first position estimation unit. In the second position estimation unit, the parameter is varied so that the difference between the first position candidate and the second position candidate is reduced (S104, S105, S106) .

  Preferably, the first position estimation unit (11) estimates the position based on an inductance in the fixed coordinate system of the motor (30) varying depending on the position, and the second position The estimation unit (12) estimates the position based on a back electromotive voltage generated by the motor.

According to a second aspect of the motor position estimating apparatus of the present invention, a first position estimating unit (11) that estimates the position of the motor (30) and outputs it as a first position candidate (θ ^ _LOW ); A second position estimation unit (12) that estimates the position and outputs it as a second position candidate (θ ^ _HI ), and compares the rotational speed (ω ^ _re ) of the motor with a predetermined threshold (C) A switching determination unit (14) that performs the switching, a switching unit (13) that selects and outputs one of the first position candidate and the second position candidate, and an output (θ of the switching unit (13)) ^ _Re ) is subjected to a low-pass filtering process and output as an estimated value (θ ^ _re ⁺ ) of the position, and the cut-off frequency of the low-pass filtering process is the result of the comparison of the switching determination unit. And low-pass transmission filtering means (16) that is variable on the basis. At a rotational speed higher than the predetermined threshold, the second method has higher accuracy of position estimation than the first method.

  Preferably, the first position estimating means (11) estimates the position based on the fact that the inductance of the fixed coordinate system of the motor (30) varies depending on the position. The second position estimating means (12) estimates the position based on a back electromotive voltage generated by the motor.

A third aspect of the motor position estimating apparatus according to the present invention is the motor position estimating apparatus according to the first to third aspects, wherein in the first position estimating means (11), the rotational speed ( A high frequency signal (v ^* _h ) having a frequency higher than ω _re ) is input to the motor, and the position of the motor is estimated based on a high frequency component of a current flowing through the motor.

The motor position estimation apparatus according to any one of the first to third aspects can be employed in an inverter control apparatus.

  According to the first aspect of the position estimation method according to the present invention, the first technique advantageous in low-speed rotation is switched to the second technique advantageous in high-speed rotation without changing the estimated position value.

  According to the second aspect of the position estimation method according to the present invention, after the first and second position candidates are made equal to each other in step (b), the first technique is switched to the second technique. The estimated position value does not fluctuate rapidly.

According to the third aspect of the position estimation method according to the present invention, the first method advantageous for low-speed rotation is changed to the second method advantageous for high-speed rotation without rapidly changing the estimated position value. Switch. Moreover, the estimated value is gradually shifted in step (b).

According to the fourth aspect of the position estimation method according to the present invention, the second method advantageous for high-speed rotation is changed to the first method advantageous for low-speed rotation without rapidly changing the estimated position value. Switch. Moreover, the estimated value is gradually shifted in step (c).

According to the fifth aspect of the position estimation method according to the present invention, from the first method advantageous in low speed rotation to the second method advantageous in high speed rotation without abruptly changing the estimated position value. Switch.

According to the sixth aspect of the position estimation method according to the present invention, the second method advantageous for high-speed rotation is changed to the first method advantageous for low-speed rotation without abruptly changing the estimated position value. Switch.

According to the seventh aspect of the position estimation method according to the present invention, hunting is unlikely to occur in the estimated position value in the vicinity of the rotational speed at which the first method and the second method are switched.

According to the eighth aspect of the position estimation method of the present invention, an estimated value of the position can be obtained even when the motor rotates at a low speed, particularly when the motor is stopped.

  According to the first aspect of the position estimation apparatus of the present invention, the motor position estimation method according to the first and second aspects of the present invention can be executed.

According to the second aspect of the position estimation apparatus of the present invention, the motor position estimation method according to the fifth aspect and the sixth aspect of the present invention can be executed.

According to the third aspect of the position estimation apparatus of the present invention, it is possible to obtain an estimated value of the position even when the motor rotates at a low speed, particularly when the motor is stopped.

First embodiment.
FIG. 1 is a block diagram showing a control system to which the position estimation technique according to the first embodiment of the present invention can be applied. The control system includes a position estimation unit 1, a speed control unit 21, a current control unit 22, a superimposition unit 23, a PWM signal generation unit 24, an inverter 25, 3/2 phase conversion units 26 and 27, a rotary coordinate conversion unit 28, a differential Device 29 and IPM motor 30.

The speed control unit 21 generates a current command value i ^* based on a difference between a rotational speed command value ω _re ^* and a rotational speed estimated value ω ^ _re described later. The rotational speed command value ω _re ^* is given from outside the control system.

The current control unit 22 generates a voltage command value v ^* based on a q-axis current i _q , a d-axis current i _d and a current command value i ^* , which will be described later. The voltage command value v ^* is composed of, for example, a q-axis voltage command value and a d-axis voltage command value.

  The speed control unit 21 and the current control unit 22 can be configured by a PI controller or the like.

The PWM signal generation unit 24 generates a switching pattern of the inverter 25 based on the voltage command value v ^* . When performing low-speed position estimation to be described later, the voltage command value v ^* frequency voltage command value v ^* _h are superimposed by the superimposing unit 23.

The inverter 25 performs switching based on the above switching pattern, and three-phase voltages v _u , v _v and v _w are applied to the IPM motor 30. At this time, currents i _u , i _v , i _w flowing in the respective phases are detected.

The 3 / 2-phase conversion unit 26 phase-converts the three-phase voltages v _u , v _v , and v _w into two-phase fixed coordinates, and outputs two-phase voltages v _α and v _β . The 3 / 2-phase conversion unit 27 converts the three-phase currents i _u , i _v , and i _w into two-phase fixed coordinates and outputs two-phase currents i _α and i _β . Rotation coordinate conversion unit 28 outputs the two-phase currents _{i α,} i β a to phase change into a rotating coordinate biphasic aforementioned q-axis current i _q, d-axis current i _d.

The position estimation unit 1 receives the two-phase voltages v _α and v _β and the two-phase currents i _α and i _β and obtains and outputs a position estimation value θ ^ _re . The rotation coordinate conversion employed in the rotation coordinate conversion unit 28 is calculated based on the position estimated value θ ^ _re .

Differentiator 29 differentiates the position estimate theta ^ _re time, outputs a rotation speed estimation value omega ^ _re described above.

The position estimation unit 1 includes a low speed position estimation unit 11, a high speed position estimation unit 12, a switching unit 13, and a switching determination unit 14. The low-speed position estimation unit 11 and the high-speed position estimation unit 12 both receive the two-phase voltages v _α and v _β and the two-phase currents i _α and i _β and estimate the positions obtained based on the respective estimation methods. The values are output as position candidates θ ^ _LOW and θ ^ _HI .

The high-speed position estimation unit 12 estimates the position based on the circuit equation of the motor 30 and the back electromotive voltage generated by the motor 30, and outputs this as a position candidate θ ^ _HI . For example, the technique described in Non-Patent Document 2 is executed. The circuit equation has a predetermined parameter, for example, q-axis inductance, and its value is variable.

The low-speed position estimation unit 11 estimates the position of the motor 30 based on, for example, the inductance variation of the motor 30, and outputs this as the position candidate θ ^ _LOW . For example, the high-frequency component of the current output from the motor by inputting a high-frequency signal using the fact that the inductance of the fixed coordinate system varies depending on the position as in the methods described in Non-Patent Documents 1 and 2. Calculate based on According to this method, an estimated value of the position can be obtained even when the motor rotates at a low speed, particularly when the motor is stopped. Non-Patent Document 2 discloses that the position estimation error does not depend on variations in inductances Lq and Ld, armature winding resistance, and field flux linkage number in the rotating coordinate system. That is, the inductance Lq in the rotating coordinate system, which is a robust parameter in the position estimation in the low-speed position estimation unit 11, affects the position estimation in the high-speed position estimation unit 12.

However, when the low-speed position estimation unit 11 is executed, the superimposition unit 23 superimposes the high-frequency voltage command value vector v ^* _h on the voltage command value vector v ^* . For example, the superimposing unit 23 may be operated by a control signal (not shown) from the low speed position estimating unit 11 so as to be performed when the low speed position estimating unit 11 operates.

The switching determination unit 14 compares the estimated rotational speed value ω ^ _re with a predetermined threshold C. As the predetermined threshold C, a value that is advantageous in that the position candidate θ ^ _HI has higher position estimation accuracy than the position candidate θ ^ _{LOW at a} higher rotational speed than this is selected. The switching means 13 selects one of the position candidates θ ^ _LOW and θ ^ _HI and outputs it as the estimated position value θ ^ _re of the position estimation unit 1. Which is output is determined by the switching determination unit 14.

The switching means 13 also compares the size of the position candidates θ ^ _LOW and θ ^ _HI . Based on the comparison result, the high-speed position estimating unit 12 increases or decreases the q-axis inductance. On the other hand, based on the comparison result, the switching determination unit 14 determines the selection of the switching means 13.

FIG. 2 is a flowchart showing the position estimation method according to this embodiment. First, in step S101, position candidates θ ^ _LOW and θ ^ _HI are calculated. As described above, these can be realized by the operation of the low-speed position estimating unit 11 and the operations of the high-speed position estimating unit 12 and the superimposing unit 23, respectively.

In step S102, it is determined whether or not the estimated rotational speed value ω ^ _re of the motor 30 is higher than a predetermined threshold C. This is executed in the switching determination unit 14. For example, the switching determination unit 14 stores a predetermined threshold value C.

For a while after the start-up, the rotation speed of the motor 30 is low, and the determination result in step S102 is negative, and the process proceeds to step S107. As a result, the switching determination unit 14 causes the switching unit 13 to adopt the position candidate θ ^ _LOW as the estimated position value θ ^ _re and outputs it from the position estimation unit 1.

As long as the situation where the estimated rotational speed value ω ^ _re is equal to or less than the predetermined threshold C continues, that is, in the situation where the estimated rotational speed value ω ^ _re never exceeds the predetermined threshold value C from the start, only the position candidate θ ^ _LOW Is adopted as the estimated value θ ^ _re of the position. Therefore, it is not necessary to operate the high-speed position estimating unit 12 under such a situation, and it is not always necessary to obtain the position candidate θ ^ _HI in step S101. When the operation of the high-speed position estimating unit 12 is omitted, the superimposing unit 23 continues to operate.

When the rotation speed of the motor 30 increases and the estimated rotation speed value ω ^ _re shifts from a value equal to or less than the predetermined threshold C to a value exceeding the threshold C, the determination result in step S102 becomes affirmative. However, this alone does not cause the switching determination unit 14 to cause the switching means 13 to adopt the position candidate θ ^ _HI as the estimated position value θ ^ _re . The switching determination unit 14 controls the selection of the switching unit 13 depending on the result of the magnitude comparison in the switching unit 13.

Proceeding from step S102 to step S103, the switching means 13 determines whether the position candidates θ ^ _HI and θ ^ _LOW are equal to each other (whether the difference is within a predetermined threshold range). Therefore, the position estimation by the high-speed position estimation unit 12 needs to be executed before reaching step S103. In making this calculation it does not operate the superimposing unit 23, the voltage command value v ^* not superimposed high-frequency voltage command value v ^* _h.

If an affirmative determination is obtained in step S103, the estimated position value θ ^ _re does not change even if the estimation method is switched. Therefore, the accuracy of the high-speed rotation executed by the high-speed position estimation unit 12 can be improved from the estimation method that is highly accurate and advantageous in the low-speed rotation executed by the low-speed position estimation unit 11 without changing the position estimation value θ ^ _re. Can be switched to a high and advantageous estimation method. Specifically, the process proceeds to step S108, and the switching determination unit 14 causes the switching unit 13 to adopt the position candidate θ ^ _HI as the position estimated value θ ^ _re .

If a negative determination is obtained in step S103, the position estimation value θ ^ _re changes if the estimation method is switched, so that the switching determination unit 14 does not cause the switching unit 13 to switch the selection. Then, a variable parameter, for example, q-axis inductance Lq, included in the circuit equation of the motor 30 is changed so that the position candidates θ ^ _LOW and θ ^ _HI are equal.

Specifically, in step S104, it is determined whether or not the position candidate θ ^ _HI is larger than the position candidate θ ^ _LOW . If the determination is affirmative, the process proceeds to step S105, and the q-axis inductance estimated value Lq ^ is decreased. This is because the position candidate θ ^ _HI increases as the estimated value Lq ^ of the q-axis inductance decreases. Conversely, if the determination in step S104 is negative, the process proceeds to step S106, and the estimated value Lq ^ of q-axis inductance is increased so as to decrease the position candidate θ ^ _HI . In step S107, the position candidate θ ^ _LOW remains adopted as the position estimation value θ ^ _re . Thereafter, if the position candidates θ ^ _HI and θ ^ _LOW are equal to each other, the process proceeds from step S103 to step S108, and the switching determination unit 14 estimates the position candidate θ ^ _HI for the switching means 13 from the position. The value θ ^ _re is adopted.

As described above, according to the present embodiment, the position estimation method is switched after the position candidates θ ^ _LOW and θ ^ _HI are equal to each other, so that the position estimation method can be switched without fluctuation of the estimated value.

Second embodiment.
FIG. 3 is a block diagram showing a control system to which the position estimation technique according to the second embodiment of the present invention can be applied. The control system has a configuration in which the switching means 13 is replaced with a calculation means 15 with respect to the control system shown in the first embodiment.

The calculation means 15 calculates an interpolation value between the position candidates θ ^ _HI and θ ^ _LOW . Then, based on the comparison result of the switching determination unit 14, the interpolation value or one of the position candidates θ ^ _HI and θ ^ _LOW is output as the estimated position value θ ^ _re .

  FIG. 4 is a flowchart showing the position estimation method according to this embodiment. Although not shown, interpolation parameters k and g are given 0 as initial values in advance.

First, in step S201, position candidates θ ^ _HI and θ ^ _LOW are calculated in the same manner as in step S101. Next, in step S202, similarly to step 102, it is determined whether or not the estimated rotational speed value ω ^ _re is higher than a predetermined threshold C.

For some time after the start-up, the rotational speed of the motor 30 is low, and the determination result of step S202 is negative, and the process proceeds to step S208. In step S208, it is determined whether the parameter k is larger than a predetermined value, for example, 1000. The parameter k has not yet been updated after startup. Therefore, the determination obtains a negative result and proceeds to step S210. In step S210, as in step S107, the switching determination unit 14 causes the calculation means 15 to employ the position candidate θ ^ _LOW as the estimated position value θ ^ _re and outputs it from the position estimation unit 1.

As long as the situation where the estimated rotational speed value ω ^ _re is equal to or less than the predetermined threshold C continues, that is, in the situation where the estimated rotational speed value ω ^ _re never exceeds the predetermined threshold value C from the start, only the position candidate θ ^ _LOW Is adopted as the estimated value θ ^ _re of the position. Therefore, it is not necessary to operate the high-speed position estimating unit 12 under such a situation, and it is not always necessary to obtain the position candidate θ ^ _HI in step S201. When the operation of the high-speed position estimating unit 12 is omitted, the superimposing unit 23 continues to operate.

When the rotation speed of the motor 30 increases and the estimated rotation speed value ω ^ _re shifts from a value equal to or less than the predetermined threshold C to a value exceeding the threshold C, the determination result in step S202 becomes affirmative. However, this alone does not cause the switching determination unit 14 to cause the calculation means 15 to adopt the position candidate θ ^ _HI as the estimated position value θ ^ _re . Until the predetermined time elapses, the switching determination unit 14 causes the calculation means 15 to adopt the interpolated value as the estimated position value θ ^ _re . A value that gradually shifts from the position candidate θ ^ _LOW to the position candidate θ ^ _HI during the predetermined period is adopted as the estimated value θ ^ _re .

Proceeding from step S202 to step S203, it is determined whether or not the parameter k is larger than a predetermined value, for example, 1000, as in step S208. Immediately after the rotational speed estimated value ω ^ _re exceeds the threshold C from a value equal to or less than the predetermined threshold C, the determination result is negative and the process proceeds to step S205. Here, the calculation means 15 calculates an interpolation value, for example, a linear interpolation value (k · θ ^ _HI + (1000-k) · θ ^ _LOW ) / 1000. Then, the interpolation value is output from the position estimation unit 1 as the estimated value θ ^ _re . Thereafter, in step S206, the parameter k is updated to increase by a predetermined increment, for example, 1.

Thereafter, if the rotational speed estimated value ω ^ _re is in a state where the parameter k and g are not initialized and the rotational speed estimated value ω ^ _re exceeds the threshold C, steps S201, S205, and S206 are repeated, and the estimated value θ ^ _re gradually increases. Thus, the position candidate θ ^ _LOW is shifted to the position candidate θ ^ _HI . When the parameter k becomes 1001 due to the update in step S206, the process proceeds from step S203 to step S204. In step S204, the switching determination unit 14 causes the calculation means 15 to adopt the position candidate θ ^ _HI as the estimated position value θ ^ _re .

  As described above, when the rotational speed increases, the estimation method is switched after the position candidates gradually shift, so that the estimation method can be switched without abruptly changing the position estimation value.

Even if the estimated rotational speed value ω ^ _re once exceeds the threshold value C for a predetermined period or more and then the estimated rotational speed value ω ^ _re becomes equal to or less than the threshold value C again, the process proceeds from step S202 to step S208. move on. However, the value of the parameter k has already exceeded 1000, and the determination result in step S208 is affirmative. However, this alone does not cause the switching determination unit 14 to cause the calculation means 15 to adopt the position candidate θ ^ _LOW as the position estimation value θ ^ _re . Until the predetermined time elapses, the switching determination unit 14 causes the calculation means 15 to adopt the interpolated value as the estimated position value θ ^ _re . A value that gradually shifts from the position candidate θ ^ _HI to the position candidate θ ^ _LOW in the predetermined period is adopted as the estimated value θ ^ _re .

In step S208, a positive determination result is obtained, and the process proceeds to step S209. In step S209, it is determined whether or not the parameter g is larger than a predetermined value, for example, 1000. Immediately after the estimated rotational speed value ω ^ _re falls below the threshold C from a value _{exceeding the} predetermined threshold C, the determination result is negative, and the process proceeds to step S211. Here, the calculation means 15 calculates an interpolation value, for example, a linear interpolation value (g · θ ^ _LOW + (1000-g) · θ ^ _HI ) / 1000. Then, the interpolation value is output from the position estimation unit 1 as the estimated value θ ^ _re . Thereafter, in step S212, the parameter g is updated to increase by a predetermined increment, for example, 1.

Thereafter, if the rotation speed estimated value ω ^ _re is in a state equal to or smaller than the threshold C without initialization of the parameters k and g, steps S201, S211 and S212 are repeated, and the estimated value θ ^ _re is gradually increased. the transition from position candidates θ ^ _HI to the position candidate θ ^ _LOW to. When the parameter g reaches 1001 due to the update in step S212, the process proceeds from step S209 to step S210. In step S210, the switching determination unit 14 causes the calculation means 15 to adopt the position candidate θ ^ _LOW as the estimated position value θ ^ _re .

  As described above, even when the rotational speed is decreased, the estimation method is switched after the position candidates gradually shift, so that the estimation method can be switched without abruptly changing the position estimation value.

  The interpolation value may be obtained by calculation other than linear interpolation, but it is desirable to adopt linear interpolation to gradually shift the estimated value so as not to increase the speed at which the estimated position value fluctuates.

Third embodiment.
FIG. 5 is a block diagram showing a control system to which the position estimation technique according to the third embodiment of the present invention can be applied. The control system has a configuration in which a low-pass filter 16 that performs a low-pass transmission filtering process is added to the output of the switching unit 13 with respect to the control system shown in the first embodiment.

  FIG. 6 is a flowchart showing a position estimation method according to the third embodiment of the present invention.

Calculation means 15 in the present embodiment, the position candidate θ ^ _HI, θ ^ instead of calculating an interpolated value between the _LOW position candidate θ ^ _HI, θ ^ low-pass filter to either the _LOW Process and output as estimated position value θ ^ _re ⁺ . The cut-off frequency of this low-pass transmission filtering process changes based on the comparison result of the switching determination unit 14.

As in the second embodiment, the interpolation parameters k and g are given 0 as initial values in advance. First, in step S301, the position candidates θ ^ _HI and θ ^ _LOW are calculated in the same manner as in step S101. Next, in step S302, similarly to step 102, it is determined whether or not the estimated rotational speed value ω ^ _re is higher than a predetermined threshold value C.

For a while after the start-up, the rotational speed of the motor 30 is low, and the determination result in step S302 is negative, and the process proceeds to step S308. In step S308, the switching unit 13 outputs the position candidate θ ^ _LOW to the low-pass filter 16 as the value θ ^ _re . In step S309, it is determined whether the parameter k is larger than a predetermined value, for example, 1000. The parameter k has not yet been updated after startup. Therefore, the determination obtains a negative result and proceeds to step 312. Then, the cut-off frequency f1 is adopted as a parameter for low-pass transmission filtering.

As long as the situation where the rotational speed estimated value ω ^ _re is below the predetermined threshold C continues, that is, in the situation where the rotational speed estimated value ω ^ _re never exceeds the predetermined threshold C from the time of start-up, low-pass transmission filtering was performed. Only the position candidate θ ^ _LOW is adopted as the position estimate θ ^ _re ⁺ . Therefore, it is not necessary to operate the high-speed position estimating unit 12 under such a situation, and it is not always necessary to obtain the position candidate θ ^ _HI in step S301. When the operation of the high-speed position estimating unit 12 is omitted, the superimposing unit 23 continues to operate.

When the rotation speed of the motor 30 increases and the estimated rotation speed value ω ^ _re shifts from a value equal to or less than the predetermined threshold C to a value exceeding the threshold C, the determination result in step S302 becomes affirmative. For the subsequent processing, the position candidate θ ^ _HI needs to be obtained.

Proceeding from step S302 to step S303, the switching means 13 outputs the position candidate θ ^ _HI as the value θ ^ _re to the low-pass filter 16. Proceeding to step S304, similarly to step S309, it is determined whether or not the parameter k is larger than a predetermined value, for example, 1000. Immediately after the estimated rotational speed value ω ^ _re exceeds the threshold C from a value equal to or less than the predetermined threshold C, the determination result is negative, and the process proceeds to step S306. Then, a cutoff frequency f2 (<f1) is adopted as a parameter for low-pass transmission filtering. In step S307, the parameter k is updated to increase by a predetermined increment, for example, 1. The order of steps S306 and S307 may be reversed.

After that, if the rotation speed estimated value ω ^ _re exceeds the threshold C without initialization of the parameters k and g, steps S301, S303, S306, and S307 are repeated, and the estimated value θ ^ _re ^The position candidate θ ^ _{HI that} has been low-pass filtered is output as ⁺ .

  When the parameter k reaches 1001 due to the update in step S307, the process proceeds from step S304 to step S305. In step S305, the cut-off frequency f3 (> f2) is adopted as a parameter for low-pass transmission filtering.

  As described above, when the rotational speed increases, when the estimation method is switched, the cutoff frequency of the low-pass transmission filter is once decreased from f1 to f2, and then increased to f3. The estimation method can be switched without abruptly changing the value.

Once the estimated rotational speed value ω ^ _re exceeds the threshold C for a predetermined period or longer, and the rotational speed estimated value ω ^ _re falls below the threshold value C again, the process proceeds from step S302 to steps S308 and S309. Advances. However, the value of the parameter k has already exceeded 1000, and the determination result in step S309 is affirmative. However, this alone does not return the cutoff frequency to the value f1. Until the predetermined time elapses, the cutoff frequency adopts a value f4 lower than the values f1 and f3.

In step S309, a positive determination result is obtained, and the process proceeds to step S310. In step S310, it is determined whether the parameter g is larger than a predetermined value, for example, 1000. Immediately after the estimated rotational speed value ω ^ _re falls below the threshold C from a value _{exceeding the} predetermined threshold C, the determination result is negative, and the process proceeds to step S311. In step S311, the cutoff frequency of the low-pass transmission filter is set to the value f4.

  In step S313, the parameter g is updated to increase by a predetermined increment, for example, 1. The order of steps S311 and S313 may be reversed.

Thereafter, if the rotation speed estimation value ω ^ _re is not more than the threshold value C without initialization of the parameters k and g, steps S301, S308, S311, and S313 are repeated. When the parameter g reaches 1001 due to the update in step S313, the process proceeds from step S310 to step S312. In step S312, the cutoff frequency of the low-pass transmission filtering is returned to the value f1.

  As described above, when the rotational speed is lowered, when the estimation method is switched, the cutoff frequency of the low-pass transmission filtering is once lowered from f3 to f4 and then raised to f1, so that the position estimation is performed. The estimation method can be switched without suddenly changing the value.

In the above description, the case where the low-pass filter processing is separately performed by the low-pass filter 16 is illustrated. However, the low-pass filter processing for the position candidates θ ^ _LOW and θ ^ _{HI is performed by} the low-speed position estimation unit 11 and the high-speed position, respectively. You may make the estimation part 12 perform. In that case, an instruction to change the cutoff frequency is given from the switching determination unit 14 to the low-speed position estimation unit 11 and the high-speed position estimation unit 12.

Other variations.
In the second and third embodiments, the same threshold value C is used for both the case where the rotational speed is increased and the case where the rotational speed is decreased, but they may be different. When switching between a method suitable for position detection at low speed rotation and a method suitable for position detection at high speed rotation, hunting is unlikely to occur in the estimated position value. For example, the first threshold value when switching from a method suitable for position detection at low speed rotation to a method suitable for position detection at high speed rotation is used as the first threshold value from the technique suitable for position detection at high speed rotation. By making the value larger than the second threshold value when switching to a method suitable for the method, hunting is unlikely to occur in the estimated position value when switching between both methods. At this time, the position estimation unit 12 for high speed can estimate the position more accurately than the position estimation unit 11 for low speed at a rotational speed higher than the first threshold, and the speed is high at a rotational speed lower than the second threshold. It is desirable to select the first threshold value and the second threshold value so that the low-speed position estimation unit 11 can estimate the position more accurately than the use position estimation unit 12.

  The position estimation method according to the present invention can be employed in the control method of the inverter 25, and the position estimation apparatus according to the present invention can be employed in the control apparatus of the inverter 25.

It is a block diagram which shows the control system which can apply the position estimation technique concerning the 1st Embodiment of this invention. It is a flowchart which shows the position estimation method concerning the 1st Embodiment of this invention. It is a block diagram which shows the control system which can apply the position estimation technique concerning the 2nd Embodiment of this invention. It is a flowchart which shows the position estimation method concerning the 2nd Embodiment of this invention. It is a block diagram which shows the control system which can apply the position estimation technique concerning the 3rd Embodiment of this invention. It is a flowchart which shows the position estimation method concerning the 3rd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Position estimation part 11 Low speed position estimation part 12 High speed position estimation part 13 Switching means 14 Switching determination part 15 Calculation means 21 Speed control part 22 Current control part 23 Superimposition part 24 PWM signal generation part 25 Inverter 26, 27 3/2 Phase converter 28 Rotating coordinate converter 29 Differentiator 30 IPM motor

Claims (18)

A method for estimating the position of the motor (30), which employs the first method (11) and the second method (12) to estimate the position of the motor (30),
The estimation of the position by the second method is affected by a robust parameter in the estimation of the position in the first method,
The position estimated by the first method is set as a first position candidate (θ ^ _LOW ),
The position estimated by the second method is set as a second position candidate (θ ^ _HI ),
(A) adopting the first position candidate as an estimated value (θ ^ _re ) of the position when the rotational speed (ω ^ _re ) of the motor is equal to or less than a predetermined threshold (C) (S107); ,
(B) When the rotation speed shifts from a value equal to or less than a predetermined threshold (C) to a value exceeding the threshold, the rotation is performed so that the second position candidate becomes equal to the first position candidate. A motor position estimation method comprising: changing the parameter (Lq ^) in the coordinate system (S105, S106) and employing the second position candidate obtained thereafter as the estimated value (S108). In the first method (11), the position is estimated based on the fact that the inductance in the fixed coordinate system of the motor (30) varies depending on the position,
The motor position estimation method according to claim 1, wherein, in the second method (12), the position is estimated based on a back electromotive voltage generated by the motor. The parameter is an estimated value (Lq ^) of the q-axis inductance of the motor,
The step (b)
When the second position candidate (θ ^ _HI ) is larger than the first position candidate (θ ^ _LOW ), increasing an estimated value of the q-axis inductance of the motor (S105);
When the second position candidate is less than or equal to the first position candidate, reducing the estimated value of the q-axis inductance (S106);
And adopting the first position candidate as an estimated value (θ ^ _re ) of the position until the second position candidate becomes equal to the first position candidate (S107). The motor position estimation method according to claim 2. A method for estimating the position of the motor (30), which employs the first method (11) and the second method (12) to estimate the position of the motor (30),
The position estimated by the first method is set as a first position candidate (θ ^ _LOW ),
The position estimated by the second method is set as a second position candidate (θ ^ _HI ),
(A) In a situation where the rotational speed (ω ^ _re ) of the motor never exceeds the first threshold (C) from the time of startup, the first position candidate is adopted as the estimated value (θ ^ _re ) of the position Performing step (S210);
(B) When the rotational speed shifts from a value equal to or less than the first threshold value to a value exceeding the first threshold value,
(B-1) adopting, as the estimated value, a value that gradually shifts from the first position candidate to the second position candidate (S205);
(B-2) After the second position candidate is adopted as the estimated value, the step of continuing to adopt the second position candidate as the estimated value (S204),
At a rotational speed higher than the first threshold, the second method has a higher estimation accuracy of the position than the first method,
In the step (b-1),
The estimated value is a value that is linearly interpolated between the first position candidate and the second position candidate as time elapses, and from the first position candidate (θ ^ _LOW ) to the second position candidate. A motor position estimation method that linearly shifts to a candidate (θ ^ _HI ). In the first method (11), the position is estimated based on the fact that the inductance in the fixed coordinate system of the motor (30) varies depending on the position,
The motor position estimation method according to claim 4, wherein, in the second method (12), the position is estimated based on a back electromotive voltage generated by the motor. A method for estimating the position of the motor (30), which employs the first method (11) and the second method (12) to estimate the position of the motor (30),
The position estimated by the first method is set as a first position candidate (θ ^ _LOW ),
The position estimated by the second method is set as a second position candidate (θ ^ _HI ),
(A) In a situation where the rotational speed (ω ^ _re ) of the motor never exceeds the first threshold (C) from the time of startup, the first position candidate is adopted as the estimated value (θ ^ _re ) of the position Performing step (S210);
(B) When the rotational speed shifts from a value equal to or less than the first threshold value to a value exceeding the first threshold value,
(B-1) adopting, as the estimated value, a value that gradually shifts from the first position candidate to the second position candidate (S205);
(B-2) After the second position candidate is adopted as the estimated value, the step of continuing to adopt the second position candidate as the estimated value (S204),
At a rotational speed higher than the first threshold, the second method has a higher estimation accuracy of the position than the first method,
(C) When the rotational speed (ω ^ _re ) shifts from a value higher than the second threshold value (C) to a value equal to or lower than the second threshold value,
(C-1) adopting a value that gradually shifts from the second position candidate (θ ^ _HI ) to the first position candidate (θ ^ _LOW ) as the estimated value (S211);
(C-2) further comprising a step (S210) of continuously adopting the first position candidate as the estimated value after the first position candidate is adopted as the estimated value;
At a rotational speed lower than the second threshold, the first method has a higher estimation accuracy of the position than the second method,
In step (c-1),
The estimated value is a value that is linearly interpolated between the second position candidate and the first position candidate as time elapses, and is calculated from the second position candidate (θ ^ _HI ). A method for estimating the position of a motor, which linearly changes to a candidate (θ ^ _LOW ). A method for estimating the position of the motor (30), which employs the first method (11) and the second method (12) to estimate the position of the motor (30),
The position estimated by the first method is set as a first position candidate (θ ^ _LOW ),
The position estimated by the second method is set as a second position candidate (θ ^ _HI ),
(A) In a situation where the rotational speed (ω ^ _re ) of the motor never exceeds the first threshold (C) from the time of startup, the first position candidate (θ ^ _LOW ) is set to the first cutoff frequency ( a step (S308, S312) of performing the low-pass transmission filtering process in f1) and adopting the estimated value (θ ^ _re ⁺ ) of the position;
(B) When the rotational speed shifts from a value equal to or less than the first threshold value to a value exceeding the first threshold value,
(B-1) In a predetermined period (S304), the second position candidate (θ ^ _HI ) is subjected to low-pass transmission filtering processing at a second cutoff frequency (f2) lower than the first cutoff frequency. Step (S303, S306) employed as the estimated value of the position;
(B-2) After the predetermined period, the second position candidate is subjected to low-pass filtering with a third cutoff frequency (f3) higher than the second cutoff frequency, and the position And adopting as an estimated value (S303, S305),
The motor position estimation method in which the second method has a higher estimation accuracy of the position than the first method at a rotational speed higher than the first threshold. In the first method (11), the position is estimated based on the fact that the inductance in the fixed coordinate system of the motor (30) varies depending on the position,
The motor position estimation method according to claim 7, wherein in the second method (12), the position is estimated based on a back electromotive voltage generated by the motor. (C) When the rotational speed of the motor shifts from a value higher than a second threshold value to a value equal to or lower than the second threshold value,
(C-1) In a predetermined period, the first position candidate (θ ^ _LOW ) is low-pass filtered at a fourth cutoff frequency (f4) lower than the third cutoff frequency (f3). Processing and adopting (S308, S311) as an estimated value (θ ^ _re ) of the position;
(C-2) After the predetermined period, the first position candidate is subjected to low-pass transmission filtering processing at the first cut-off frequency (f1) and adopted as an estimated value of the position (S308, S312). And a step of
9. The motor position according to claim 7, wherein the first method has a higher estimation accuracy of the position than the second method at a rotational speed lower than the second threshold. Estimation method.   The motor position estimation method according to claim 6, wherein the first threshold is higher than the second threshold. In the first method, a high-frequency signal (v * _h ) having a frequency higher than the rotational speed (ω _re ) is input to the motor, and the position of the motor is estimated based on a high-frequency component of a current flowing through the motor. The motor position estimation method according to any one of claims 1 to 10.   An inverter control method using the motor position estimation method according to any one of claims 1 to 11. A first position estimator (11) that estimates the position of the motor (30) and outputs it as a first position candidate (θ ^ _LOW );
A second position estimation unit (12) that estimates the position and outputs the second position candidate (θ ^ _HI );
A switching determination unit (14) for comparing the rotational speed (ω ^ _re ) of the motor with a predetermined threshold (C);
Switching means (13) for outputting one of the first position candidate and the second position candidate as an estimated value (θ ^ _re ) of the position based on the comparison result of the switching determination unit; With
The estimation of the position in the second position estimation unit is affected by a robust parameter (Lq ^) in the estimation of the position in the first position estimation unit,
In the second position estimation unit, the parameter is varied so that a difference between the first position candidate and the second position candidate is reduced (S104, S105, S106), and a motor position estimation device (1 ). In the first position estimating unit (11), the position in the fixed coordinate system of the motor (30) is estimated based on the fluctuation depending on the position,
The motor position estimation device according to claim 13, wherein the second position estimation unit (12) estimates the position based on a back electromotive voltage generated by the motor. A first position estimator (11) that estimates the position of the motor (30) and outputs it as a first position candidate (θ ^ _LOW );
A second position estimation unit (12) that estimates the position and outputs the second position candidate (θ ^ _HI );
A switching determination unit (14) for comparing the rotational speed (ω ^ _re ) of the motor with a predetermined threshold (C);
Switching means (13) for selecting and outputting one of the first position candidate and the second position candidate;
The output (θ ^ _re ) of the switching means (13) is subjected to a low-pass transmission filtering process and output as an estimated value (θ ^ _re ⁺ ) of the position, and the cut-off frequency of the low-pass transmission filtering process is Low pass transmission filtering means (16) that is variable based on the comparison result of the switching determination unit , and
The motor position estimation apparatus in which the second method has higher accuracy of position estimation than the first method at a rotational speed higher than the predetermined threshold. In the first position estimating means (11), the position of the fixed coordinate system of the motor (30) is estimated based on fluctuations depending on the position,
The motor position estimating device according to claim 15, wherein the second position estimating means (12) estimates the position based on a back electromotive voltage generated by the motor. In the first position estimating means (11) , a harmonic signal (v ^* _h ) having a frequency higher than the rotational speed (ω _re ) is input to the motor, and based on the high frequency component of the current flowing through the motor. The motor position estimation apparatus according to any one of claims 13 to 16, wherein the motor position is estimated. An inverter control device using the motor position estimation device according to any one of claims 13 to 17 .

Images