JP4576857B2 - Rotor position estimation method, motor control method, and program - Google Patents

Description

The present invention, the rotor position estimation how the permanent magnet motor, a motor control method, and a program.

  In a permanent magnet motor (DC brushless motor, IPM motor) having saliency, the rotor position is calculated from voltage, current information, etc. obtained on the drive inverter side, and the motor is operated without using a physical position sensor or speed sensor. A so-called sensorless control technique for controlling is used.

  As sensorless control methods, a rectangular wave sensorless control method and a sine wave sensorless control method are known. The rectangular wave sensorless control method is a method for detecting an induced voltage, and the technology has already been established. However, in this method, the driving waveform is limited to rectangular wave driving centering on 120 ° driving.

  Aiming for low noise and high efficiency, there is a great need to make the drive waveform of the IPM motor a sine wave and to secure reliability and reduce costs by position sensorless drive. The sine wave sensorless control method estimates a position by detecting an electric current and calculating an induced voltage.

ｋ_Eは、誘起電圧定数である。 In the 180 ° sine wave sensorless control, a dq axis coordinate system including a d axis that is a position in the magnetic flux direction of the permanent magnet rotor and a q axis that is orthogonal to the rotation direction from the d axis is used. The voltage equation based on the motor model in the dq axis real rotation coordinate system is shown in the following [Equation 1].

k _E is an induced voltage constant.

The motor equivalent circuit equation for the d-axis component is considered to be misaligned when there is no position sensor, and the following equation [Equation 2] is taken into consideration for the induced voltage component and Ed as follows. Become.

From the above equation [Equation 2], the induced voltage Ed is expressed as the following equation [Equation 3].

  Techno Frontier Symposium 2003 Session Sponsored by Japan Management Association Session C-5 Toshiba Production Technology Center at the Motor Technology Symposium C5-1-5 (Non-Patent Document) entitled “High-performance motor technology in air conditioners” by Mr. Sekihara Document 1) describes that the angular velocity (motor rotation speed) is determined so that the induced voltage Ed = 0 corresponding to the estimation error of the rotor position is obtained, and this value is integrated to estimate the angle (rotor position). Has been.

Toshiba Corp. Junichi Sekihara Material entitled “High-performance motor technology in air conditioners” Techno Frontier Symposium 2003 Session C-5 Motor Technology Symposium sponsored by Japan Management Association April 18, 2003

  However, the induced voltage Ed is a value proportional to the angular velocity ω (number of rotations). That is, since the induced voltage Ed is a function of the position estimation error and the rotational speed, the design of a controller such as an axis deviation adjuster is complicated, and it is difficult to obtain stability over a wide operating range. There is a problem. In particular, in the low speed range, the position estimation accuracy is poor due to the calculation accuracy and the like.

An object of the present invention is easy to design the controller is to provide stability easily obtained rotor position estimation how the control method of the motor, and a program.

  The rotor position estimation method of the present invention is a rotor position estimation method for estimating the rotor position of a permanent magnet motor having saliency, and the voltage equation of the permanent magnet motor is divided by the angular velocity ω of the rotor. The magnetic flux error between the rotational coordinate axis and the estimated axis of the rotational coordinate axis is obtained based on the magnetic flux equation, and the rotor position is estimated based on the magnetic flux error.

  In the rotor position estimation method of the present invention, the magnetic flux error is a magnetic flux error between the q-axis that is the rotational coordinate axis and the δ-axis that is the estimated axis of the rotational coordinate axis.

本発明の回転子位置推定方法は、永久磁石モータの回転子位置を推定する回転子位置推定方法であって、回転座標軸と前記回転座標軸の推定軸との磁束誤差を求め、前記磁束誤差に基づいて、前記回転子位置を推定し、前記磁束誤差は、下記式［数５］により求められることを特徴としている。

Rotor position estimation method of the present invention, there is provided a rotor position estimation method of estimating a rotor position of the permanent magnet motor, determine the flux error between the estimated axis of the the rotation axis rotating coordinate axes, the flux error based on estimates the rotor position, the magnetic flux error is characterized Rukoto be determined by the following formulas [expression 4].

A rotor position estimation method according to the present invention is a rotor position estimation method for estimating a rotor position of a permanent magnet motor, and obtains a magnetic flux error between a rotational coordinate axis and an estimated axis of the rotational coordinate axis, and is based on the magnetic flux error. Then, the rotor position is estimated, and the magnetic flux error is obtained by the following equation [Equation 5].

In the rotor position estimation method of the present invention, the magnetic flux error is obtained by the following equation [Equation 6 ].

In the rotor position estimation method of the present invention, the magnetic flux error is obtained by the following equation [Equation 7 ].

  In the rotor position estimation method according to the present invention, the inductance is a function depending on at least one of a current and a rotational speed.

  According to the motor control method of the present invention, the angular velocity of the rotor corresponding to the position error between the rotational coordinate axis corresponding to the magnetic flux error and the estimated axis of the rotational coordinate axis obtained by the rotor position estimation method of the present invention is determined. An estimated value is obtained, an estimated value of the angular velocity of the rotor is input to a low-pass filter, and feedback control relating to the speed of the permanent magnet motor is performed based on an output value from the low-pass filter.

  The motor control method of the present invention is a motor control method to which the rotor position estimation method of the present invention is applied, and a value corresponding to the detected value of the estimated axis component of the rotational coordinate axis of the armature current and a command A phase command value, which is the output of the current controller for making the value error zero, is input to the low-pass filter, and based on the output value from the low-pass filter and the estimated rotor position, the voltage command It is characterized by generating a signal indicating a phase.

  The motor control method of the present invention is a motor control method to which the rotor position estimation method of the present invention is applied, and the value corresponding to the detected value of the estimated axis component of the rotating coordinate axis of the armature current is low-passed. A voltage phase command value is generated based on a deviation between the output value from the low-pass filter and the command value of the estimated axis component of the rotational coordinate axis of the armature current.

The motor control method of the present invention is a motor control method for controlling a permanent magnet motor having saliency, and the amount of magnetic flux on the δ-axis, which is the estimated axis of the q-axis, which is the rotational coordinate axis, is estimated for the motor applied voltage. The value Vγ ^ is obtained as the sum of the magnetic flux calculated by dividing the voltage drop difference between the resistor R and the current Iγ by the angular velocity ωre ^ and the magnetic flux obtained by the product of the inductance Lq and the current Iδ . It is characterized by controlling the amount of magnetic flux to converge to zero.

The motor control method of the present invention is a motor control method for controlling a permanent magnet motor having saliency, and the amount of magnetic flux on the δ-axis, which is the estimated axis of the q-axis, which is the rotational coordinate axis, is estimated for the motor applied voltage. From the value Vγ ^, the magnetic flux calculated by dividing the difference between the voltage drop of the resistor R and the current Iγ and the voltage drop due to the temporal change of the inductance Ld and the current Iγ by the angular velocity ωre ^, the inductance Lq and the current calculated as the sum of the obtained magnetic flux by the product of i?, it is characterized that you control to converge the magnetic flux amount of the δ-axis to zero. According to the present invention, responsiveness can be improved.

The motor control method of the present invention is a motor control method for controlling a permanent magnet motor having saliency, and the amount of magnetic flux on the δ-axis, which is the estimated axis of the q-axis, which is the rotational coordinate axis, is estimated for the motor applied voltage. Magnetic flux calculated from the value Vγ ^ by dividing the difference between the voltage drop of the resistor R and the current Iγ and the voltage drop caused by the product of the positive gain constant K "and the current Iγ over time by the angular velocity ωre ^ When, determined as the sum of the obtained magnetic flux by the product of the inductance Lq and the current it?, the δ according to the magnetic flux amount to control so as to converge to zero is characterized in Rukoto. the present invention of the shaft, as Ld By not using a motor constant, the system configuration is simplified. When K ″ is zero, the system is the same as the above invention.

  The program of this invention is a program for making a computer perform each step of the rotor position estimation method of the said invention.

  According to the present invention, the controller can be easily designed and stability can be easily obtained. Further, the position estimation accuracy in the low speed region is improved.

  Hereinafter, an embodiment of a rotor position estimation method of the present invention will be described in detail with reference to the drawings.

  This embodiment relates to 180 ° sine wave sensorless control of a permanent magnet motor (DC brushless motor, IPM motor) having saliency. As shown in FIG. 6, in the 180 ° sine wave sensorless control, the dq axis actual consisting of the d axis which is the position of the permanent magnet rotor in the magnetic flux direction and the q axis which is advanced 90 ° in the rotational direction from the d axis. The axis deviation Δθ between the rotation coordinate system, the control virtual rotor position γ-axis, and the δ-δ axis consisting of the δ-axis advanced by 90 ° from the γ-axis is obtained, and the Δθ is made zero. Control.

  The rotor angle θ (rotor position) is obtained by integrating the angular velocity ω of the rotor. Therefore, when controlling the positional deviation (Δθ) to zero, Δθ becomes zero based on Δθ. ω is obtained (speed estimator 24 in FIG. 3 described later), and the motor is controlled based on the ω.

  In this embodiment, as will be described later, the positional deviation Δθ between the q-axis and the δ-axis is calculated from the estimated magnetic flux error, and when the phase of the δ-axis is a leading phase, the estimated angular velocity on the δ-axis is reduced. If the phase on the δ axis is a delayed phase, the estimated angular velocity on the δ axis is increased, and the position θ corresponding to the integrated value of Δθ is made to coincide with the q axis.

  Here, when Δθ is obtained as a value proportional (dependent) to ω, it becomes difficult to design a controller on the rear stage side where Δθ is input. In gain design of a controller, etc., the dependency of the rotational speed (ω) tends to be strong, and there is a problem that it is difficult to adjust the gain, and the convergence state of the estimated value of Δθ depends on the speed. This adjustment also requires experimentation and complexity. For this reason, when Δθ is determined as a value that depends on ω, it is difficult to ensure stability and a control error may occur.

  On the other hand, the position error estimator (see reference numeral 21 in FIG. 3) of the present embodiment uses a magnetic flux vector instead of using an induced voltage vector, as will be described later, so that Δθ depends on ω. It can be obtained as a value not to be Thereby, in this embodiment, the design of the controller (speed estimator 24) to which Δθ is input is easy (usually a general PI controller can be used), and sufficient stability can be secured. The occurrence of control errors is suppressed.

  Since the magnetic flux of the permanent magnet is obtained by integrating the induced voltage over time, the magnetic flux vector is estimated at a position delayed by 90 ° from the induced voltage vector, and the magnitude is the amount of magnetic flux (physical quantity) centered on the permanent magnet. )become. Therefore, it is possible to directly estimate the q-axis by estimating the electrical absolute position using a magnetic flux vector having a physical quantity that does not vary with the rotational speed. As a result, since the design of the estimator for calculating the estimated speed is based on the magnetic flux error that does not depend on the rotational speed, the adjustment of the parameters becomes easy.As a result, in the calculation of the output of the speed controller and the estimated position, Stability is improved, and motor drive can be easily stabilized.

  With reference to FIG. 3, a motor control apparatus to which the rotor position estimation method of the present embodiment is applied will be described.

  In FIG. 3, the motor control device 10 sensorlessly drives a permanent magnet motor 20 having saliency. The motor control device 10 includes a PWM inverter 17, a current detector (not shown), coordinate transformation calculation units 27 and 28, a position error estimator 21, a speed estimator 24, an integrator 26, and a speed control. And a voltage controller 15, a voltage generator 15, and a voltage compensator 16.

The PWM inverter 17 converts a DC voltage into a three-phase AC voltage.
The current detector (not shown) detects the current i (u, w) of the motor 20.
The coordinate conversion calculation unit 27 converts the detected current i (u, w) into a rotation coordinate.
The position error estimator 21 estimates the rotor position error Δθ ^.
The speed estimator 24 estimates the angular speed ωre ^ so that the estimated position error Δθ ^ becomes zero.
The integrator 26 integrates the output ωre ^ of the speed estimator 24 to calculate the rotor position θ (γ, δ) ^.

The speed controller 12 estimates the rotor position, and is used to make an error between the speed command value ωre ^* and the output ωre ^ of the speed estimator 24 zero.
The current controller 14 is used to make the error between the γ-axis (d-axis) current command value iγ ^* and the γ-axis current iγ obtained from the actually detected current information zero.
The PWM output value V (u, v, w) ^* is calculated based on the voltage command value V (m) ^* output from the speed controller 12 and the phase command value Vβ ^* output from the current controller 14. .
The voltage generator 15 outputs to the PWM inverter 17 based on the voltage amplitude command V (m) ^* output from the speed controller 12 and the phase V (θ) ^{* of the} voltage command output from the adder 18. A voltage command V (u, v, w) is generated.
The voltage compensator 16 inputs the voltage command value V (u, v) ^* to the PWM inverter 17 generated by the voltage generator 15 and corrects the phase and amplitude with respect to the value V (u, v) ^* . The estimated voltage value Vmd (u, v) ^ is output.
The coordinate transformation calculation unit 28 transforms the estimated voltage value Vmd (u, v) ^ output from the voltage compensator 16 into rotational coordinates.

  3 to 5 show different configuration examples of the motor control device 10 of the present embodiment. The difference between FIGS. 3, 4 and 5 differs depending on which output stability is improved, and a low-pass filter (LPF) is inserted into the control output to be stabilized. Although the responsiveness may slightly decrease due to the insertion of the LPF, the LPF is inserted when securing the stability is given priority. Which configuration of FIGS. 3 to 5 is adopted depends on the characteristics of the motor and the characteristics of the load. Other than the configuration related to the LPF, this is a common system configuration. Hereinafter, the configuration of the motor control device 10 will be described with reference to FIG. 3, and FIG. 4 or FIG. 5 will be referred to as necessary.

The command values for the motor control device 10 are the angular velocity ωre ^* and the γ-axis current Iγ ^* .
The adder 11 calculates a deviation between the angular velocity command value ωre ^* and the estimated angular velocity value ωre ^. The deviation is input to a speed controller 12 constituted by a PI (proportional integral) controller. The output command output from the speed controller 12 is the voltage amplitude command V (m) ^* . This voltage amplitude command V (m) ^* is an amplitude command of the three-phase command voltage of the motor 20.

The adder 13 calculates a deviation between the γ-axis current command value Iγ ^* and the Iγ detected and calculated from the motor current. The deviation is input to the current controller 14 configured by the PI controller. The output command output from the current controller 14 is a voltage phase command value Vβ ^* . This voltage phase command value Vβ ^* is a phase command of the three-phase command voltage of the motor 20. As shown in FIG. 5, in order to stabilize the operation of the motor 20, the output of the command value Vβ ^* of the voltage phase is passed through the digital LPF 31 to reduce the vibration component and the like, thereby stabilizing the command voltage phase. May be.

The adder 18 calculates the sum of the voltage phase command Vβ ^* and the estimated rotor position θ (γδ) ^ as the voltage command phase V (θ) ^* . The phase V (θ) ^* of the voltage command is input to the voltage generator 15. In the voltage generator 15, for example, the following command voltage waveform is generated.

  Such a voltage command V (u, v, w) is output to the PWM inverter 17. The PWM inverter 17 is configured by an inverter circuit or the like, and generates a PWM waveform. Since this PWM inverter 17 is a general PWM inverter that has been frequently used in the past, the description thereof is omitted.

  When the IPM motor 20 is actually driven by the PWM inverter 17, the phase current of the motor 20 is detected. If this detection circuit uses a CT or the like and an amplifier circuit using an operational amplifier is configured on the secondary side of the CT, a value (waveform) obtained by easily converting the phase current into a voltage signal can be obtained. Since the waveform of the motor phase current is an analog value, it is converted into a digital value by an AD converter or the like and converted into a value that can be calculated.

Further, since the phase currents iu and iw of the motor are currents as seen from the stationary coordinate system, the coordinate transformation calculation unit 27 performs coordinate transformation on the estimated rotational coordinate system. This transformation matrix is the following matrix.

  Since the calculated currents iγ and iδ are superimposed with harmonic components and noise caused by the motor 20, the calculation results are passed through a digital LPF 32 (FIG. 5) to reduce the harmonic components and noise. can do. The LPF 32 can be omitted depending on the characteristics of the motor 20 and the characteristics of the inverter device 17 and the like.

In the adder 13, a deviation between the current iγ and the γ-axis current command value iγ ^* is calculated, and the deviation is input to the current controller 14. Further, the current iγ is used for calculation in the position error estimator 21 and calculation in the inductance compensator 22. On the other hand, the current iδ is used for calculation in the position error estimator 21 and calculation in the inductance compensator 22.

  The inductance compensation in the inductance compensator 22 is configured to compensate for inductance saturation and modeling error. This inductance compensation is configured to be a function of at least one variable of current (iγ, iδ) and rotational speed. For this compensation, an approximate expression or a table may be used.

The voltage compensator 16 inputs the voltage command value V (u, v) ^* to the PWM inverter 17 generated by the voltage generator 15 and corrects the phase and amplitude with respect to the value V (u, v) ^* . The estimated voltage value Vmd (u, v) ^ is output. The correction performed by the voltage compensator 16 takes into consideration the nonlinearity of input / output in the PWM inverter 17 and is performed so as to correspond to the output from the PWM inverter 17 to the motor 20.

  Hereinafter, the calculation in the position error estimator 21 will be described in detail.

ｋ_Eは、誘起電圧定数である。 The voltage equation based on the motor model in the dq axis real rotation coordinate system is shown in the following equation [Equation 10 ].

k _E is an induced voltage constant.

The voltage equation based on the motor model in the γ-δ axis estimated rotational coordinate system is expressed by the following equation [Equation 11 ].

Here, each parameter of the inductance L is expressed by the following formula [Equation 12 ].

Here, since Δθ is controlled to be zero, Δθ≈0.
sin Δθ≈0 and cosΔθ≈1. When this approximation is used, the following formula [Equation 13 ] is obtained for Δθ.

Here, since the induced voltage V is obtained by differentiating the magnetic flux φ, the following formula [Formula 14 ] is obtained as a relational expression regarding the amount of magnetic flux. That is, by dividing the voltage equation of [Equation 13 ] by the angular velocity ωre, the following [Equation 14 ] magnetic flux equation is obtained.

ここで、電流の過渡項は無視し、ｐ（ｉ）≒０とした。また、Ｋ’は、誘起電圧定数の逆数に相当し、Ｋ’＞０を満たす任意の定数または、関数のように可変にさせてもよい。 Using the approximation of Δθre≈sin Δθre, the following equation [Equation 15 ] is obtained.

Here, the transient term of the current is ignored and p (i) ≈0. K ′ corresponds to the reciprocal of the induced voltage constant, and may be made variable as an arbitrary constant or function satisfying K ′> 0.

Since the DC motor rotates, V has a value proportional to ω. Therefore, in the equation [ 15 ], Vγ is a value proportional to ω. Therefore, the term of (Vγ−Riγ) / ωre in the formula [ 15 ] does not depend on ω. From this, stable control can be performed by obtaining Δθre from the above [Equation 15 ] and performing control to make Δθre zero.

Ｋ”はＫ”＞０を満たす任意の定数または関数として制御可能である。応答性を重視する場合には、Ｋ”を大きな値とし、応答性よりも安定性を重視したい場合には、Ｋ”＝０とする。この場合には、上記［数１５］と同じ式となる。 Further, when p (i) ≠ 0 and the transient term is taken into consideration, the following equation [Equation 16 ] is obtained.

K ″ can be controlled as any constant or function that satisfies K ″> 0. In the case where importance is attached to responsiveness, K ″ is set to a large value, and in the case where stability is more important than responsiveness, K ″ = 0 is set. In this case, the equation is the same as the above [Equation 15 ].

FIG. 1 is a block diagram showing a configuration of a position error estimator 21 that performs position error estimation using the above equation [Expression 15 ]. FIG. 2 is a block diagram showing the configuration of the position error estimator 21 that performs position error estimation using the above equation [Equation 16 ]. In FIG. 2, the transient term corresponds to (Δiγ / ωre). As shown in FIGS. 1 and 2, the position error estimator 21 includes a magnetic flux calculator that estimates the magnetic flux φδ and a position error calculator that estimates the position error Δθ based on the estimated magnetic flux φδ. . 1 and 3, R represents a winding resistance of the motor. The q-axis inductance Lq ^* is obtained by the inductance compensator 22. The q-axis inductance Lq ^* is a function of iγ, iδ, and ωre obtained in advance by experiments using a motor with a sensor.

  In FIG. 1, the magnetic flux calculation unit of the position error estimator 21 integrates the difference (Vγ−R · Iγ) of the voltage drop R · Iγ in the current from the motor applied voltage Vγ ((Vγ−R · Iγ)). / Ωre) and a magnetic flux (Lq · iδ) obtained by multiplying the motor winding inductance Lq and the current iδ, and rotation based on the estimated magnetic flux ((Vγ−R · Iγ) / ωre + Lq · iδ). The magnetic flux error φδ between the coordinate axis (q axis) and the estimated axis (δ axis) of the rotational coordinate axis is obtained, and the position error calculation unit of the position error estimator 21 estimates the rotor position Δθ based on the magnetic flux error φδ. To do.

In FIG. 2, the magnetic flux calculator of the position error estimator 21 integrates the difference (Vγ− (R + pLd) · Iγ) of the voltage drop ((R + pLd) · Iγ) in the current from the motor applied voltage Vγ. ((Vγ− (R + pLd) · Iγ) / ωre) and the product of the inductance Lq of the motor winding and the current iδ (Lq · iδ) are estimated, and the estimated magnetic flux ((Vγ− (R + pLd) · Based on (Iγ) / ωre + Lq · iδ), a magnetic flux error φδ between the rotational coordinate axis (q axis) and the estimated axis (δ axis) of the rotational coordinate axis is obtained, and the position error calculation unit of the position error estimator 21 Based on the error φδ, the rotor position Δθ is estimated.
As shown in FIG. 3, the position error Δθre (Δθ ^) obtained by the position error estimator 21 is output to the speed estimator 24.

The position error Δθ ^ calculated by the position error estimator 21 is input to a speed estimator 24 configured by a PI controller. The speed estimator 24 calculates an angular speed estimated value ωre ^ such that Δθ ^ becomes zero. Here, as the speed estimator 24, a general PI controller is usually used. The calculation formula in the speed estimator 24 is as shown in the following formula [Equation 17 ].

The estimated angular velocity value ωre ^ obtained by the velocity estimator 24 is output to the adder 11 and used for velocity feedback control as described above. A voltage amplitude command V (m) ^* is generated by the speed feedback control. Further, the estimated angular velocity value ωre ^ is integrated by the integrator 26 to calculate the rotor position estimated value θre (γδ) ^. The calculated position estimation value θre (γδ) ^ is input to the voltage phase command adder 18 and the coordinate converters 27 and 28, respectively.

The adder 18 generates a voltage phase command V (θ) ^* input to the voltage generator 15 based on the position estimated value θre (γδ) ^. As described above, the positional deviation of Δθ ^ estimated by the position error estimator 21 is reflected in the voltage phase command V (θ) ^* , so that the positional deviation of Δθ ^ is reflected in the motor 20.

  The angular velocity estimated value ωre ^ is adjusted so that the actual position of the motor 20 matches the estimated value, and the position θre (γδ) ^ on the estimated coordinate system, which is an integral value thereof, is obtained. Feedback control is performed so that the positions of the γ-δ axes coincide. Further, when the estimated angular velocity value ωre ^ is used for velocity feedback control, the LPF 33 (FIG. 4) may be passed to stabilize it.

  As described above, in the motor control device 10 of the present embodiment, the value detected by the sensor is used as the current value, but the command value or the estimated value is used as the voltage value (the voltage sensor is not being used). An inductance estimated value (command value) is obtained by the inductance compensator 22 for the inductance, which is an important factor in the operation of the motor having the saliency (IPM motor). Regarding the angular velocity ω, in the present embodiment, an estimated value is used in consideration of the fact that a transient one is represented, but a command value can also be used.

As described above, according to the present embodiment, a physical quantity unrelated to ωre can be obtained by performing position estimation using magnetic flux. By using the equation of [Equation 15 ] or [Equation 16 ], gain adjustment is facilitated, and motor control with improved stability can be realized by simple and short-time calculation. If the motor is controlled by the motor control apparatus to which the rotor position estimation method of the present embodiment is applied, stable motor operation can be achieved. If a motor controlled by a motor control device to which the rotor position estimation method of this embodiment is applied is used as a compressor motor, a highly efficient and low noise compressor (not shown) can be provided. Moreover, if the compressor is applied to an air conditioner (not shown), it can contribute to reduction of power consumption of the air conditioner.

  Compared with the technique described in Non-Patent Document 1, since the induced voltage is a value proportional to the rotational speed (ω), the position error Δθ estimated based on the induced voltage is a value dependent on ω, While the possibility of a large deviation from the true value is high, the magnetic flux of the permanent magnet is a material-specific value that does not depend on ω, so the position error Δθ estimated based on the magnetic flux is true or true. The value is close to the value. Therefore, the magnetic flux is suitable for use in position estimation (position error estimation).

It should be noted that when obtaining Δθre, the following equations [Equation 18 ] and [Equation 19 ] can be developed from the equation [Equation 13 ].

Here, in order to perform the control of Δθ≈0, the control can be performed by the following equation [Equation 20 ] considering only the numerator without actually considering the denominator of the above equation [Equation 19 ]. is there.

In order to reduce the amount of calculation in order to perform control for Δθ≈0 in real time, Δθ is obtained from an equation approximated by sin as in the equation of [Equation 20 ] instead of the equation of [Equation 19 ]. There is.

However, in Δθ = tan ⁻¹ () in the equation of [Equation 19 ], the ω dependency is well offset (since there is a term that depends on ω in the denominator and the numerator, both can be canceled and the ω dependency is On the other hand, when calculating the equation [ 20 ], Δθ becomes a value dependent on ω.

This is because the induced voltage is calculated in the equation [ 20 ], which is a feature amount proportional to the rotational speed, and the above-described equation [3] is used (the technology of the non-patent document 1). ) Will have the same problem. That is, the convergence of Δθ to 0 is poor, and gain adjustment of the axis deviation adjuster requires different gain adjustment depending on the rotation speed of the motor 20, which complicates the control configuration. On the other hand, according to the present embodiment, as described above, it is possible to estimate Δθ unrelated to ω by performing position estimation using magnetic flux.

  In the motor control device 10 of the present embodiment, it is preferable to obtain Δθ (magnetic flux error) between the q axis and the δ axis. A magnetic flux φ by a permanent magnet exists on the d-axis, and φ = 0 on the q-axis orthogonal to the d-axis. When Δθ is obtained between the q axis and the δ axis, feedback control may be performed so that φ = 0 (q axis). This is easier to control than in the case of obtaining Δθ between the d-axis and the γ-axis and performing feedback control to φ (d-axis). When Δθ is obtained between the d axis and the γ axis and feedback control is performed on the magnetic flux φ, the magnetic flux φ is a value inherent to the material of the permanent magnet. This is because the command value must be changed for each motor. On the other hand, instead of the above, also in this embodiment, Δθ can be obtained between the d-axis and the γ-axis.

It is a block diagram of a calculation when there is no transient term in the position error estimator of the motor control device to which an embodiment of the rotor position estimation method of the present invention is applied. It is a block diagram of a calculation when there exists a transient term in the position error estimator of the motor control apparatus to which an embodiment of the rotor position estimation method of the present invention is applied. It is a block diagram which shows the structure of the motor control apparatus with which one Embodiment of the rotor position estimation method of this invention is applied. It is a block diagram which shows the other structure of the motor control apparatus with which one Embodiment of the rotor position estimation method of this invention is applied. It is a block diagram which shows further another structure of the motor control apparatus with which one Embodiment of the rotor position estimation method of this invention is applied. It is a figure which shows the model for sensorless control.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Motor controller 11 Adder 12 Speed controller 13 Adder 14 Current controller 15 Voltage generation part 16 Voltage compensator 17 PWM inverter 21 Position error estimator 22 Inductance compensator 24 Speed estimator 26 Integrator 27 Coordinate conversion calculation part 28 Coordinate transformation calculation unit 31 LPF
32 LPF
33 LPF

Claims (14)

A rotor position estimation method for estimating a rotor position of a permanent magnet motor having saliency ,
Based on the magnetic flux equation obtained by dividing the voltage equation of the permanent magnet motor by the angular velocity ω of the rotor, the magnetic flux error between the rotational coordinate axis and the estimated axis of the rotational coordinate axis is obtained,
A rotor position estimation method, wherein the rotor position is estimated based on the magnetic flux error. Te rotor position estimating method smell of claim 1,
The rotor position estimation method , wherein the magnetic flux error is a magnetic flux error between a q-axis that is the rotation coordinate axis and a δ-axis that is an estimation axis of the rotation coordinate axis . A rotor position estimation method of estimating a rotor position of the permanent magnet motor,
Seek flux error of the rotating coordinate axes and the estimated axis of the rotation axis,
Based on the magnetic flux error, estimate the rotor position ,
The magnetic flux error is obtained by the following equation [Equation 1].

And a rotor position estimation method. Met rotor position estimation method of estimating a rotor position of the permanent magnet motor,
Obtain the magnetic flux error between the rotational coordinate axis and the estimated axis of the rotational coordinate axis,
Based on the magnetic flux error, estimate the rotor position,
The magnetic flux error is obtained by the following equation [Equation 2].

And a rotor position estimation method. Claim 1 or 2 rotor position estimating method smell according to Te,
The magnetic flux error is obtained by the following equation [Equation 3].

And a rotor position estimation method. In the rotor position estimation method according to claim 1 or 2 ,
The magnetic flux error is obtained by the following equation [Equation 4 ].

And a rotor position estimation method. The rotor position estimation method according to any one of claims 3 to 6 ,
The method of estimating a rotor position , wherein the inductance is a function depending on at least one of a current and a rotational speed . Angular velocity of the rotor that corresponds to the position error between the estimated axis of the rotation axis and the rotation coordinate axis corresponding to the more the obtained magnetic flux error in the rotor position estimating method according to any one of claims 1 7 Find an estimate of
The estimated value of the angular velocity of the rotor is input to a low-pass filter,
On the basis of the output value from the low-pass filter, the control method of the motor, characterized in that feedback control on the velocity of the permanent magnet motor is Ru performed. A motor control method to which the rotor position estimation method according to any one of claims 1 to 7 is applied,
The value corresponding to the detected value of the estimated axis component of the rotational coordinate axis of the armature current and the phase command value that is the output of the current controller for making the error of the command value zero are input to the low-pass filter,
Wherein the output value from the low-pass filter, on the basis of the said estimated rotor position, the control method of the motor, characterized in that that generates a signal indicative of the phase of the voltage command. A motor control method to which the rotor position estimation method according to any one of claims 1 to 7 is applied,
Enter a value corresponding to the detected value of the estimated axis component of the rotational axis of the armature current to the low-pass filter,
A motor phase control value is generated based on a deviation between an output value from the low-pass filter and a command value of an estimated axis component of a rotation coordinate axis of the armature current . A motor control method that controls a permanent magnet motor having saliency,
The amount of magnetic flux on the δ axis, which is the estimated axis of the q axis, which is the rotational coordinate axis,
Magnetic flux calculated by dividing the difference between the voltage drop of the resistor R and the current Iγ from the estimated value Vγ ^ of the motor applied voltage by the angular velocity ωre ^;
Obtained as the sum of the magnetic flux obtained by the product of the inductance Lq and the current Iδ,
A motor control method, wherein control is performed so that the amount of magnetic flux of the δ axis converges to zero . A motor control method for controlling a permanent magnet motor having saliency,
The amount of magnetic flux on the δ axis, which is the estimated axis of the q axis, which is the rotational coordinate axis ,
From the estimated value Vγ ^ of the motor applied voltage, the magnetic flux calculated by dividing the difference between the voltage drop of the resistor R and the current Iγ and the voltage drop due to the temporal change of the inductance Ld and the current Iγ by the angular velocity ωre ^ ,
Obtained as the sum of the magnetic flux obtained by the product of the inductance Lq and the current Iδ,
A motor control method, wherein control is performed so that the amount of magnetic flux of the δ axis converges to zero. A motor control method for controlling a permanent magnet motor having saliency ,
The magnetic flux amount of δ-axis is an estimated axis of the q-axis is a rotation axis,
From the estimated value of the motor applied voltage V.gamma ^, the voltage drop across the resistor R and the current i?, The positive gain constant K "a difference between the voltage drop due to the product of the temporal change of the current i?, Is divided by the angular velocity? Re ^ And the magnetic flux calculated by
Obtained as the sum of the magnetic flux obtained by the product of the inductance Lq and the current Iδ ,
Control method of the motor, characterized in that that control so as to converge the magnetic flux amount of the δ-axis to zero. The program for making a computer perform each step of the rotor position estimation method of any one of Claim 1 to 7 .

Images