JP2010028981A - Rotor position estimating method for synchronous motor, and controller for the synchronous motor - Google Patents

Abstract

An object of the present invention is to enable position estimation with high resolution and high accuracy even during high-speed rotation or when an induced voltage increases.
A method for estimating a rotor position of a synchronous motor according to the present invention includes: (a) a phase voltage value of a stator winding, a phase current value of a stator winding, a resistance of a stator winding, and a stator winding; A step of estimating the induced voltage of each phase of the synchronous motor based on the inductance; and (b) creating an induced voltage calculation value obtained by coordinate-converting the induced voltage of each phase to a control axis rotating at an arbitrary speed; (C) creating an induced voltage reference value obtained by converting the reference value of the induced voltage on the stationary coordinate to the control axis; (d) creating a deviation between the induced voltage calculated value and the induced voltage reference value; e) estimating the rotor position by correcting the position of the control axis so that the deviation converges to zero.
[Selection] Figure 4

Description

  The present invention relates to a method for controlling a synchronous motor, and more particularly to a method for estimating a rotor position of a synchronous motor. Furthermore, the present invention relates to a synchronous motor control device employing a position sensorless system.

  Since a motor called a brushless DC motor is a synchronous motor, position information of the rotor is required for control. In particular, in order to drive a motor with high efficiency in a wide speed range, it is necessary to control a current vector based on highly accurate position information, and it is usual to drive in a closed loop. If a rotor position sensor is provided to obtain high-accuracy position information, the number of wiring by the position sensor increases, the maintainability decreases, the reliability decreases, and the environment resistance when used in a special atmosphere such as in a compressor This causes problems such as a decrease in size, a size expansion based on the physique of the sensor itself, and cost. In order to eliminate this drawback, a position sensorless control device that estimates the rotor position without using a sensor has been proposed.

  Conventional position sensorless control devices can be broadly classified into those using a magnetic saliency and those using a speed electromotive force induced in a motor winding. The latter makes it difficult to estimate the position because no speed electromotive force is generated at the time of stop and extremely low speed, but has the characteristics applicable to both salient pole type and non-salient pole type at medium and high speeds. As a conventional position sensorless control device using speed electromotive force, the one described in Patent Document 1 is known. The position sensorless control device described in Patent Document 1 will be briefly described below.

  According to the position sensorless control device of Patent Document 1, first, a reference value em (induced voltage reference value) of an induced voltage of a certain phase is created in a stationary coordinate. Further, an induced voltage is calculated from the phase current value and the phase voltage value to determine an induced voltage value e (induced voltage calculated value), a deviation ε between the induced voltage reference value em and the induced voltage calculated value e is determined, The rotor position is estimated by matching the phases. A specific method for matching the phases will be described below.

  For example, when the phase of the U-phase induced voltage calculation value (U-phase induced voltage value eu) does not match the phase of the U-phase induced voltage reference value (U-phase induced voltage reference value eu), these deviations (U-phase The deviation εu) is not zero. Therefore, the phases are matched by correcting the estimated rotor position θm so that the U-phase deviation εu converges to zero. However, since accurate estimation cannot be performed in the phase range where the U-phase induced voltage value eu and the U-phase induced voltage reference value eu cross, the phase to be estimated depends on the estimated rotor position θm or the induced voltage calculation value e. select. In Patent Document 1, as shown in FIG. 11, the phase to be estimated is switched every 60 °. Thereby, the phase difference in a sine wave can be obtained as the deviation of the DC component. The estimation accuracy is improved by always estimating the angle using the phase in which the influence of the phase difference has the greatest influence on the deviation.

In the method of Patent Document 1, an induced voltage calculation value is created based on the phase voltage equation of each phase of the stator winding. Since the stator winding voltage equation uses instantaneous values, if the phase current and the phase voltage are not sinusoidal, for example, the estimated rotor position θm can be estimated even if the phase voltage is saturated. Further, when calculating the induced voltage calculation value, the induced voltage constant is not used because the phase voltage value and the phase current value are used. In other words, since it is not affected by changes in the induced voltage constant, the amount of magnetic flux of the permanent magnet changes as the motor temperature rises due to the heat generated during motor driving or the atmosphere. The position can be estimated correctly.
Japanese Patent No. 3419725

  Since the technique of Patent Document 1 is based on three-phase stationary coordinates, the induced voltage calculation value and the induced voltage reference value are each an AC amount. Therefore, when the deviation ε between the induced voltage calculation value and the induced voltage reference value is obtained, even when there is a certain phase difference, the deviation ε includes a ripple caused by comparing sine waves in principle. In addition, since the angle is always estimated using the phase in which the influence of the phase difference has the greatest influence on the deviation, the ripple included in the deviation ε also increases. Further, before and after the phase switching to be estimated, the deviation ε includes a ripple based on the switching. In addition, when driving a synchronous motor with a high induced voltage or when driving at a high speed, the induced voltage to be compared increases, so the ripple included in the deviation ε increases, and the estimated speed calculated from the result has an influence of the ripple. It appears that the estimated speed itself pulsates and may affect the control.

  In view of the above problems, an object of the present invention is to provide a method and a control device that can perform high-resolution and high-precision position estimation even at the time of high-speed rotation or when the induced voltage becomes large without impairing the advantages of the prior art. And

  The present inventor has paid attention to not using a method of comparing alternating current amounts in order to improve the estimation accuracy of the rotor position.

That is, the present invention
Estimating the induced voltage of each phase of the synchronous motor based on the phase voltage value of the stator winding, the phase current value of the stator winding, the resistance of the stator winding, and the inductance of the stator winding When,
Creating an induced voltage calculation value obtained by coordinate-transforming the induced voltage of each phase to a control axis rotating at an arbitrary speed;
Creating an induced voltage reference value obtained by coordinate-converting the reference value of the induced voltage on the stationary coordinate to the control axis;
Creating a deviation between the induced voltage calculation value and the induced voltage reference value;
Estimating the rotor position by correcting the position of the control shaft so that the deviation converges to zero;
A rotor position estimation method for a synchronous motor is provided.

In order to carry out the above method, the present invention provides:
A voltage command calculating unit that calculates and outputs a voltage command corresponding to a deviation between the current command and the detected current using an estimated position of the rotor so that a detected current follows the current command;
A current sensor for detecting a phase current value of a synchronous motor to which a three-phase AC voltage reflecting the voltage command is applied;
An inverter that converts a DC voltage into the three-phase AC voltage according to the voltage command and applies it to the synchronous motor;
Estimating the induced voltage of each phase of the synchronous motor based on the phase voltage value of the stator winding, the phase current value of the stator winding, the resistance of the stator winding, and the inductance of the stator winding Using the current estimated position of the rotor, an induced voltage calculation value obtained by converting the induced voltage of each phase into a DC component, and an induced voltage reference value on the estimated position, and the induced voltage calculated value and A rotor position estimating unit that creates a deviation from the induced voltage reference value and corrects the estimated position so that the deviation converges to zero, thereby estimating a rotor position;
A control device for a synchronous motor is provided.

In another aspect, the present invention provides:
Creating a motor terminal voltage calculation value obtained by coordinate-converting the phase voltage of each phase of the synchronous motor to a control axis rotating at an arbitrary speed;
Creating a motor terminal voltage reference value obtained by converting the reference value of the motor terminal voltage on the stationary coordinate to the control axis; and
Creating a deviation between the motor terminal voltage calculation value and the motor terminal voltage reference value;
Estimating the rotor position by correcting the position of the control shaft so that the deviation converges to zero;
A rotor position estimation method for a synchronous motor is provided.

In order to carry out the above method, the present invention provides:
A voltage command calculating unit that calculates and outputs a voltage command corresponding to a deviation between the current command and the detected current using an estimated position of the rotor so that a detected current follows the current command;
A current sensor for detecting a phase current value of a synchronous motor to which a three-phase AC voltage reflecting the voltage command is applied;
An inverter that converts a DC voltage into the three-phase AC voltage according to the voltage command and applies it to the synchronous motor;
A motor in which the phase voltage of each phase of the synchronous motor is coordinate-converted to a control axis that rotates at an arbitrary speed and a motor terminal voltage calculation value is created, and a motor terminal voltage reference value on a stationary coordinate is coordinate-converted to the control axis A terminal voltage reference value is created, a deviation between the motor terminal voltage calculation value and the motor terminal voltage reference value is created, and the estimated position is corrected so that the deviation converges to zero, thereby estimating the rotor position. A rotor position estimating unit,
A control device for a synchronous motor is provided.

  According to the present invention, a voltage calculation value is obtained from the phase voltage equation of each phase of the stator winding, and a phase deviation between the voltage calculation value and the voltage reference value can be extracted as a DC component at a certain speed. The coordinates are converted to the rotating control axis. Then, the deviation of the DC component depending on the position error on the control axis is obtained, and the estimated rotor position (θm) is corrected so that this deviation converges to zero. In principle, since the estimated position is derived as a direct current component, ripples resulting from comparing sine waves are not superimposed on the obtained estimated position, and the controllability is excellent. In addition, since it is not necessary to use a motor constant such as an induced voltage constant in the estimation algorithm, position estimation can be realized even when the phase voltage is saturated, and even when the motor constant changes depending on the motor usage status. Highly accurate position estimation with resolution can be realized. Further, since it is not necessary to switch the phase to be estimated, ripples based on the switching can be removed, thereby improving the position estimation accuracy.

  The deviation of the DC component of the voltage calculation value that can be extracted on the control axis depends on the position error between the estimated rotor position and the actual rotor position on the control axis. From the position error, it can be determined whether the estimated rotor position is advanced or delayed from the actual rotor position in the rotation direction of the rotor. If it is advanced, the frequency of the applied voltage and / or the torque command is corrected so as to delay the estimated rotor position. If it is delayed, the frequency of the applied voltage and / or the torque command is corrected to advance the estimated rotor position.

  In one aspect of the present invention, the deviation between the voltage calculation value generated on the control axis and the voltage reference value can be extracted as DC information. Therefore, by selecting only information necessary for estimating the rotor position, the rotor position can be easily determined. Can be estimated.

  In this specification, the term “error” and the term “deviation” are used interchangeably.

  First, a rotor position estimation method will be described including a process of deriving a component depending on the position error Δθ from a voltage equation of a permanent magnet synchronous motor (PMSM). Then, the control apparatus of the synchronous motor which employ | adopted those methods is demonstrated.

<< 1. Rotor position estimation method using induced voltage >>
(Definition of coordinate axes)
First, the coordinate system will be described. FIG. 3 is an analysis model diagram of the permanent magnet synchronous motor. For simplicity of explanation, an interior permanent magnet synchronous motor (IPMSM) having two magnetic poles is shown. The d-axis and the q-axis are axes based on the actual rotor phase θ. Hereinafter, the dq coordinate axes are simply referred to as real axes. The d-axis is set in the same direction as the magnetic flux generated by the permanent magnet disposed on the rotor, and the q-axis is set in a direction advanced by 90 ° with respect to the d-axis. The phase between the stator winding and the d axis is defined as the rotor position θ.

  In FIG. 3, the counterclockwise direction is assumed to be normal rotation. The direction of the forward rotation is a direction in which the current flowing through the stator windings u, v, and w changes in the order of the u phase, the v phase, and the w phase. The dc axis and the qc axis are axes determined by the estimated rotor position θm (estimated position). Hereinafter, the dc-qc coordinate axis is referred to as a control axis. In other words, the coordinates of the control axis are estimated coordinates that the control device (FIG. 1) that executes control of the synchronous motor based on the method recognizes as the rotor position. Further, the axis rotated by the estimated rotor position θm from the stator winding u is defined as the dc axis, and the qc axis is set to be advanced by 90 ° with respect to the dc axis. Further, a difference between the rotor position θ and the estimated rotor position θm is defined as a position error Δθ (= θ−θm).

  FIG. 3 shows a state where there is a positive position error Δθ. However, when there is no error in position estimation and the position error Δθ is 0, the estimated rotor position θm and the rotor position θ match, and the d-axis The dc axis coincides, and the q axis and the qc axis coincide. In the following description, the rotor position θ, the estimated rotor position θm, and the position error Δθ are represented by electrical angles. Hereinafter, unless otherwise specified, a value related to a phase (angle) is represented by an electrical angle. Here, the mechanical angle represents the angle of the rotor itself, and (electrical angle) = (p / 2) (mechanical angle). P is the number of magnetic poles.

(Derivation of induced voltage calculation value esd (edc, eqc))
Next, a method for deriving the induced voltage calculation value esd (the induced voltage calculation value edc on the dc axis and the induced voltage calculation value eqc on the qc axis) will be described in detail.

  First, an induced voltage calculation value es (u-phase induced voltage calculation value eu, v-phase induced voltage calculation value ev, w-phase induced voltage calculation value ew) of each phase is created from a known phase voltage equation of a synchronous motor. Specifically, the induced voltage calculation value es for each phase is created based on the equations (1), (2), and (3). Here, d / dt represents time differentiation. As dθ / dt appearing in the differential calculation regarding the trigonometric function, a value obtained by converting the estimated speed ωm into the electrical angular speed is used. D (iu) / dt, d (iv) / dt, and d (iw) / dt are obtained by first-order Euler approximation. Specifically, di / dt = {i (n) −i (n−) where ΔT is the position estimation period, i (n) is the current phase current value, and i (n−1) is the previous phase current value. 1) Approximate by} / ΔT. R is a resistance per phase of the stator winding, la is a leakage inductance per phase of the stator winding, La is an average value of effective inductance per phase of the stator winding, and Las is a value of the stator winding. Represents the amplitude of effective inductance per phase. The w-phase current value iw is obtained as a value obtained by changing the sign of the sum of the u-phase current value iu and the v-phase current value iv (iw = − (iu + iv)).

  (1) The induced voltage calculation value es of each phase obtained by the equations (2) and (3) can be simplified. Specifically, assuming that the phase current values iu, iv, and iw are sine waves, the current command amplitude ia (the absolute value of the current vector) and the current command phase βT (the advance angle of the current vector from the q axis) and Phase currents iu, iv, and iw are created and substituted into equations (1), (2), and (3), respectively. Furthermore, when the equations (1), (2), and (3) are expressed using the maximum amplitude value If of the phase current and simplified, equations (4), (5), and (6) are obtained. Values obtained from current sensors may be used as the phase current values iu, iv, iw. The maximum amplitude value If of the phase current can be derived based on the value obtained from the current sensor.

  As described above, assuming that the phase current flowing in the stator winding is a sine wave, the voltage drop based on the resistance of the stator winding and the inductance of the stator winding is subtracted from the phase voltage of the stator winding. Thus, the induced voltage of each phase can be estimated. Even if the induced voltage calculation value es is obtained using the simplified phase voltage equation, the same results and effects as those obtained when the equations (1), (2), and (3) are used are obtained. By using a simplified phase voltage equation, the calculation time can be shortened.

  Since the induced voltage calculation value of each phase obtained by the equations (1), (2), (3) or (4), (5), and (6) is defined on the stationary coordinates, it is an AC amount. In the present invention, these are converted into a direct current amount by mapping them to the control axis (dc-qc coordinate axis). Specifically, coordinate conversion of the induced voltage calculation value es (u-phase induced voltage calculation value eu, v-phase induced voltage calculation value ev, w-phase induced voltage calculation value ew) of each phase is performed based on the equation (7). Thus, an induced voltage calculation value esd (dc axis induced voltage calculation value edc, qc axis induced voltage calculation value eqc) on the dc-qc axis is derived.

(Derivation of induced voltage reference value esdm (edcm, eqcm))
Next, a method for deriving the reference induced voltage reference value esdm on the dc-qc axis (the induced voltage reference value edcm on the dc axis, the induced voltage reference value eqcm on the qc axis) will be described in detail.

  First, a method for obtaining the induced voltage amplitude estimation value em will be described. Based on the result obtained by adding the absolute values of the induced voltage values of the respective phases, an induced voltage amplitude calculation value ec is created. Specifically, as in equation (8), the sum of the absolute value of the u-phase induced voltage calculation value eu, the absolute value of the v-phase induced voltage calculation value ev, and the absolute value of the w-phase induced voltage calculation value ew. Is multiplied by a certain set coefficient KEC as an induced voltage amplitude calculation value ec. Here, the coefficient KEC is given by equation (9) and is multiplied to convert the sum of the absolute values of the induced voltage values of each phase into amplitude, assuming that the induced voltage of each phase is a sine wave. Note that θm% (π / 3) is a remainder when the estimated rotor position θm (°) is divided by π / 3 (60 °).

  Further, an induced voltage amplitude estimated value em is obtained by applying a primary digital low-pass filter (LPF) to the induced voltage amplitude calculation value ec obtained by the equation (8). Specifically, the induced voltage amplitude estimated value em is obtained based on the equation (10). Here, em (n) is an estimated voltage amplitude estimated value in the current control cycle, and em (n-1) is an induced voltage amplitude estimated value in the previous control cycle. KLEM is a coefficient of the low-pass filter and takes a value from 0 to 1, and the effect of the low-pass filter increases as the value decreases. In equation (10), an error (amplitude error) between the induced voltage amplitude calculation value ec and the previous induced voltage amplitude estimated value em (n-1) is obtained, and the value obtained by multiplying the amplitude error by the coefficient KLEM is compared with the previous value. The induced voltage amplitude estimated value em (n-1) is added to the induced voltage amplitude estimated value em (n) of this time. Thus, by using the low-pass filter, the amplitude error is calculated, and the induced voltage amplitude estimated value em (n) is corrected so that the amplitude error becomes small.

  Next, an induced voltage reference value esdm that is a reference value of the induced voltage at the estimated rotor position θm is created. First, as in the equations (11), (12), and (13), the induced voltage amplitude estimated value em described above is used as the amplitude value, and the u-phase induced voltage reference value (u-phase induced voltage reference value eum) An induced voltage reference value (v-phase induced voltage reference value evm) and a w-phase induced voltage reference value (w-phase induced voltage reference value ewm) are obtained. Assuming that the permanent magnet of the rotor is sinusoidally magnetized, the induced voltage reference value esm of each phase is a sine wave.

  (11) (12) The induced voltage reference value esm of each phase determined by the equations (13) is defined on the stationary coordinates, and thus is an AC amount. Therefore, in the present invention, these are converted to a direct current amount by mapping them to the control axis. Based on the equation (14), coordinate conversion is performed on the induced voltage reference value esm (u-phase induced voltage reference value eum, v-phase induced voltage reference value evm, w-phase induced voltage reference value ewm) of each phase, so that dc−qc An induced voltage reference value esdm (dc axis induced voltage reference value edcm, qc axis induced voltage reference value eqcm) on the axis is derived.

(Derivation of deviation)
Next, a deviation ε between the induced voltage calculation value esd and the induced voltage reference value esdm is created. As in the equation (15), the difference ε is obtained by subtracting the induced voltage reference value esdm from the induced voltage calculation value esd.

  Based on the equation (15), a deviation εdc on the dc axis and a deviation εqc on the qc axis are derived. When the actual rotor position θ matches the estimated rotor position θm, the induced voltage calculation value esd and the induced voltage reference value esdm match, and the deviation becomes zero. However, if the induced voltage generated on the real axis having the actual rotor position θ is mapped to the control axis of the estimated rotor position θm when rotating with a position error Δθ, the actual rotor position θ and the estimated rotor position A deviation ε depending on the phase difference (position error Δθ) from θm occurs. This makes it possible to estimate the rotor position without the position sensor of the present invention.

(Derivation of estimated rotor position θm of rotor)
Derivation of the estimated rotor position θm is realized by correcting the estimated rotor position θm so that the deviation ε obtained by the equation (15) is converged to zero. The derivation method will be specifically described. For example, as shown in FIG. 3, when the estimated rotor position θm and the actual rotor position θ match, the dc-axis induced voltage calculation value edc and the dc-axis induced voltage reference value edcm are equal, and the qc axis The induced voltage calculation value eqc is equal to the qc-axis induced voltage reference value eqcm. However, when the estimated rotor position θm is delayed from the actual rotor position θ, a negative value depending on the position error Δθ is output as the deviation εdc on the dc axis. Further, a negative value depending on the position error Δθ is output as the deviation εqc on the qc axis. Therefore, if these conditions are satisfied, it is determined that the estimated rotor position θm is delayed, and the estimated rotor position θm is corrected to advance. Next, when the estimated rotor position θm is ahead of the actual rotor position θ, a positive value depending on the position error Δθ is output as the deviation εdc on the dc axis. Further, a negative value depending on the position error Δθ is output as the deviation εqc on the qc axis. Therefore, if these conditions are satisfied, it is determined that the estimated rotor position θm is advanced, and the estimated rotor position θm is corrected to be delayed. This makes it possible to estimate the angle with high resolution and high accuracy.

  Alternatively, the estimated rotor position θm may be corrected so that the deviation εdc (magnetic flux direction component) converges to 0, focusing only on the deviation εdc on the dc axis. In other words, the deviation εqc on the qc axis need not be derived. In this way, the calculation time can be further shortened.

  In the above-described method, the fact that the deviation ε depends on the position error Δθ is used. However, as shown in the equation (16), the position error Δθm is obtained as the deviation ε so that the deviation ε converges to zero. The estimated rotor position θm may be corrected.

<< 2. Estimation method using motor voltage >>
Next, another method for obtaining the DC component deviation ε depending on the position error Δθ will be described. In the estimation method using the above-described induced voltage, the dc-qc coordinate based on the estimated rotor position θm is used as the control axis. On the other hand, in this estimation method, γ-δ coordinates based on the motor terminal voltage position θv are used as the control axis. That is, the γ-δ coordinates are the coordinates of the motor terminal voltage that rotates with a certain phase at the same angular velocity as the estimated coordinates at which the control device that executes the control of the synchronous motor recognizes the position of the rotor based on the method. is there.

(Derivation of motor terminal voltage calculation value vsd (vγ, vδ))
Hereinafter, a method for deriving the motor terminal voltage calculation value vsd (γ-axis motor terminal voltage calculation value vγ, δ-axis motor terminal voltage calculation value vδ) will be described in detail.

First, the motor terminal voltage calculation value vs (phase u motor terminal voltage calculation value vu, v phase motor terminal voltage calculation value vv, w phase motor terminal voltage calculation value vw) of each phase from the known phase voltage value of the synchronous motor. Create As the motor terminal voltage calculation value vs of each phase, as shown in the equation (17), a voltage command vs ^* (vu ^* , vv ^* , vw ^* ) generated by a voltage command calculation unit 109 (FIG. 1 or the like) described later is used. Use.

  Since the motor terminal voltage calculation value vs of each phase obtained by the equation (17) is defined on the stationary coordinates, it is an AC amount. Therefore, in the present invention, these are converted into a direct current amount by mapping them to the control axis (γ-δ coordinate axis).

  Next, a method for deriving the motor terminal voltage position θv will be described. First, on the dc-qc axis, the voltage equation is expressed as equation (18). When the dc-axis current value idc and the qc-axis current value iqc are controlled, the dc-axis voltage value vdc and the qc-axis voltage value vqc are uniquely determined. Therefore, the voltage phase βv is also uniquely determined as shown in Equation (19). . Here, the relationship between the motor terminal voltage position θv, the voltage phase βv, and the estimated rotor position θm is expressed by the equation (20).

  By converting the motor terminal voltage calculation value vs (u-phase motor terminal voltage calculation value vu, v-phase motor terminal voltage calculation value vv, w-phase motor terminal voltage calculation value vw) of each phase by the equation (21), γ The motor terminal voltage calculation value vsd on the −δ axis (γ axis motor terminal voltage calculation value vγ, δ axis motor terminal voltage calculation value vδ) is derived.

(Derivation of motor terminal voltage reference value vsdm (vγm, vδm))
Next, a method for deriving the motor terminal voltage reference value vsdm on the reference γ-δ axis (γ-axis motor terminal voltage reference value vγm, δ-axis motor terminal voltage reference value vδm) will be described in detail.

  First, a method for obtaining the motor terminal voltage amplitude estimated value vm will be described. Based on the result obtained by adding the absolute values of the motor terminal voltage values of the respective phases, the motor terminal voltage amplitude calculation value vc is created. As in equation (22), the sum of the absolute value of the u-phase motor terminal voltage calculation value vu, the absolute value of the v-phase motor terminal voltage calculation value vv, and the absolute value of the w-phase motor terminal voltage calculation value vw A product of a certain set coefficient KEC is used as a motor terminal voltage amplitude calculation value vc. Here, the coefficient KEC is given by the equation (23), and is multiplied to convert the sum of the absolute values of the motor terminal voltages of the respective phases into the amplitude, assuming that the motor terminal voltages of the respective phases are sine waves. . Note that θv% (π / 3) is a remainder when the motor terminal voltage position θv (°) is divided by (π / 3) (60 °).

  Further, the motor terminal voltage amplitude estimated value vm is obtained by applying a first-order digital low-pass filter (LPF) to the motor terminal voltage amplitude calculated value vc obtained by the equation (22). Specifically, the estimated motor terminal voltage amplitude value vm is obtained based on the equation (24). Here, vm (n) is a motor terminal voltage amplitude estimated value in the current control cycle, and vm (n−1) is a motor terminal voltage amplitude estimated value in the previous control cycle. KLEM is a coefficient of the low-pass filter and takes a value from 0 to 1, and the effect of the low-pass filter increases as the value decreases. In equation (24), an error (amplitude error) between the motor terminal voltage amplitude calculation value vc and the previous motor terminal voltage amplitude estimated value vm (n-1) is obtained, and a value obtained by multiplying this amplitude error by a coefficient KLEM. And the previous motor terminal voltage amplitude estimated value vm (n-1) are added to obtain the current motor terminal voltage amplitude estimated value vm (n). Thus, by using the low-pass filter, the amplitude error is calculated, and the motor terminal voltage amplitude estimated value vm (n) is corrected so that the amplitude error becomes small.

  Next, a motor terminal voltage reference value vsdm, which is a reference value of the motor terminal voltage at the motor terminal voltage position θv, is created. First, as in the equations (25), (26), and (27), the motor terminal voltage amplitude estimated value vm described above is used as the amplitude value, the u-phase motor terminal voltage reference value (u-phase motor terminal voltage reference value vum), A v-phase motor terminal voltage reference value (v-phase motor terminal voltage reference value vvm) and a w-phase motor terminal voltage reference value (w-phase motor terminal voltage reference value vwm) are obtained. Assuming that the permanent magnet of the rotor is sine wave magnetized, the motor terminal voltage reference value vsm of each phase is a sine wave.

  (25) (26) The motor terminal voltage reference value vsm of each phase obtained by the equations (27) is defined on the stationary coordinates, and thus is an AC amount. Therefore, in the present invention, these are converted to a direct current amount by mapping them to the control axis. Based on the equation (28), by coordinate-transforming the motor terminal voltage reference value vsm (u-phase motor voltage reference value vum, v-phase motor terminal voltage reference value vvm, w-phase motor terminal voltage reference value vwm) of each phase, The motor terminal voltage reference value vsdm (γ-axis motor terminal voltage reference value vγm, δ-axis motor terminal voltage reference value vδm) on the γ-δ axis can be derived.

(Derivation of deviation)
Next, a deviation ε between the motor terminal voltage calculation value vsd and the motor terminal voltage reference value vsdm is created. As shown in the equation (29), a deviation ε is obtained by subtracting the motor terminal voltage reference value vsdm from the motor terminal voltage calculation value vsd.

  Based on the equation (29), a deviation εγ on the γ-axis and a deviation εδ on the δ-axis are derived. When the actual rotor position θ matches the estimated rotor position θm, the motor terminal voltage calculation value vsd matches the motor terminal voltage reference value vsdm, and the deviation is zero. However, if the motor terminal voltage generated on the real axis having the actual rotor position θ is mapped to the control axis that is the motor terminal voltage position θv when rotating with the position error Δθ, the actual rotor position θ and the estimated rotation A deviation ε depending on the phase difference (position error Δθ) of the child position θm occurs. This makes it possible to estimate the rotor position without the position sensor of the present invention.

(Derivation of estimated rotor position θm of rotor)
The derivation of the estimated rotor position θm can be realized in the same manner as the estimation method using the induced voltage.

  Alternatively, the estimated rotor position θm may be corrected so that the deviation εγ converges to 0 by paying attention only to the deviation εγ on the γ axis. In other words, it is not necessary to derive the deviation εδ on the δ axis. In this way, the calculation time can be further shortened.

  In the above-described method, the fact that the deviation ε depends on the position error Δθ is used. However, as shown in the equation (30), the position error Δθm is obtained as the deviation ε so that the deviation ε converges to zero. The estimated rotor position θm may be corrected.

  Next, a specific configuration of a synchronous motor control apparatus capable of executing the above-described method will be described.

<< first embodiment >>
As shown in FIG. 1, the synchronous motor control device 100 includes a drive unit 104, two current sensors 105 a and 105 b, a two-axis current conversion unit 106, a rotor position rotational speed estimation unit 107, a voltage command calculation unit 109, A current control unit 110, a current command generation unit 111, and a rotation speed control unit 112 are provided.

  The driving unit 104 may be a three-phase switching circuit using a switching element such as a MOSFET (Metal Oxide Semiconductor Field Effect Transistor) or an IGBT (Insulated Gate Bipolar Transistor). The other control units described above can be provided by a control application executed in a DSP (Distal Signal Processor) or a microcomputer. The DSP or microcomputer may include peripheral devices such as a core, a memory, an A / D conversion circuit, and a communication port. Of course, some of the other control units described above may be configured by a logic circuit.

  The synchronous motor 102 to be controlled is, for example, a permanent magnet synchronous motor such as IPMSM or SPMSM (Surface Permanent Magnet Synchronous Motor). The IPMSM has a saliency in which the d-axis inductance Ld and the q-axis inductance Lq are different (generally, a reverse saliency of Lq> Ld), and can use reluctance torque in addition to magnet torque, so extremely high drive efficiency Can be achieved.

  In the present embodiment, the position of the rotor is estimated without using a position sensor such as a Hall element or a resolver, and the synchronous motor 102 is controlled. In this specification, the d axis of the synchronous motor 102 recognized by the control device 100 is expressed as a dc axis, and the q axis is expressed as a qc axis.

(Operation of motor controller)
An outline of the operation of the control device 100 when the synchronous motor 102 is driven will be described. As shown in FIG. 1, a current command I ^* (torque command) is generated by the rotation speed control unit 112 based on information on the target rotation speed ω ^* given from the outside, and a dc-qc axis is generated by the current command generation unit 111. Current commands idc ^* and iqc ^* are generated. The phase currents iu and iv obtained by the current sensors 105a and 105b are converted into detected currents idc and iqc on the dc-qc coordinates by the biaxial current converter 106. Based on the deviation between the detected currents idc and iqc and the current commands idc ^* and iqc ^* , the current control unit 110 creates the voltage commands vdc ^* and vqc ^* . Based on the voltage commands vdc ^* , vqc ^* , the voltage command calculation unit 109 creates three-phase voltage commands vu ^* , vv ^* , vw ^* . The drive unit 104 generates a three-phase AC voltage having a duty corresponding to the three-phase voltage commands vu ^* , vv ^* , and vw ^* , and outputs the three-phase AC voltage to the synchronous motor 102. In this way, the synchronous motor 102 is controlled so that the actual rotational speed ω follows the target rotational speed ω ^* .

(Driver 104)
As shown in FIG. 2, the drive unit 104 includes a conversion circuit, a base driver 116, in which switching elements 113a, 113b, 113c, 113d, 113e, 113f and free-wheeling diodes 114a, 114b, 114c, 114d, 114e, 114f are paired. A smoothing capacitor 117 and a DC power supply 118. Power supply to the synchronous motor 102 is performed from the DC power supply 118 via the switching elements 113a to 113f. The DC power supply 118 corresponds to an output rectified by a diode bridge or the like. The switching pattern signal generated by the voltage command calculation unit 109 is converted into a drive signal for electrically driving the switching elements 113a to 113f by the base driver 116, and the switching elements 113a to 113f are converted into drive signals according to these drive signals. Operate.

(Rotation speed control unit 112)
The operation of the control device 100 will be described in more detail. First, in order to realize the target rotational speed ω ^* given from the outside, the current command I ^* is controlled from the deviation from the current rotational speed ω (estimated rotational speed ωm described later) using the equation (31). Calculated by the unit 112. As a calculation method, a general PI control method is used. Here, Gpω and Giω are speed control proportional gain and integral gain, ω is the rotational speed, ω ^* is the target rotational speed, and I ^* is the current command.

I ^* = Gpω × (ω ^* −ω) + Giω × Σ (ω ^* −ω) (31)

(Current command generator 111)
Furthermore, using the current command I ^* , the current command creating unit 111 calculates the dc-axis current command idc ^* and the qc-axis current command iqc ^* using the equations (32) and (33). Here, β is a current phase.

idc ^* = I ^* × −sin (β) (32)
iqc ^* = I ^* × cos (β) (33)

(Biaxial current converter 106)
The phase currents iu and iv of the synchronous motor 102 detected by the current sensors 105a and 105b are calculated based on the equation (34) by the two-axis current converter 106 and the qc-axis detection current iqc contributing to the magnet torque of the synchronous motor 102. , The dc axis detection current idc orthogonal to it is converted into a biaxial current. Here, a value derived by the above-described method is used as the estimated rotor position θm.

(Description of current control unit 110)
Then, the current control unit 110, a given current command idc ^*, and iqc ^*, the detected current idc, using iqc, (35) (36) current command by formula idc ^*, iqc ^* to the detection current idc, iqc Control calculation is performed so that the voltage commands vdc ^* and vqc ^* are obtained. Here, Gpd and Gid are a dc-axis current control proportional gain and integral gain, respectively, and Gpq and Giq are a qc-axis current control proportional gain and integral gain, respectively.

vdc ^* = Gpd × (idc ^* −idc) + Gid × Σ (idc ^* −idc) (35)
vqc ^* = Gpq × (iqc ^* −iqc) + Giq × Σ (iqc ^* −iqc) (36)

(Voltage command calculation unit 109)
Further, the voltage command calculation unit 109 is configured to drive the synchronous motor 102 based on the voltage commands vdc ^* and vqc ^* and the rotor position θm estimated by the rotor position rotation number estimation unit 107 ( Drive signal) to the drive unit 104. Specifically, from the two-phase voltage commands vdc ^* and vqc ^* , the three-phase voltage commands vu ^* , vv ^* and vw ^* are expressed as (37) using the rotor position θm so that the output waveform becomes a sine wave. It is calculated | required by the general two-phase three-phase conversion shown by Formula.

(Driver 104)
The drive unit 104 outputs U-phase, V-phase, and W-phase voltages according to the drive signal. As a result, the synchronous motor 107 is driven at the target rotation speed (speed). That is, the switching patterns of the switching elements 113a to 113f of the drive unit 104 are obtained by using the current information of the synchronous motor 102 obtained from the current sensors 105a and 105b, the information on the estimated magnetic pole position of the synchronous motor 102, and the estimation of the synchronous motor 102. It is determined from the information on the rotation speed thus obtained and the information on the target rotation speed given from the outside.

(Rotor position rotational speed estimation unit 107)
Next, details of the operation of the rotor position rotational speed estimation unit 107 will be described. As shown in FIG. 4, the rotor position rotation number estimation unit 107 is activated every set period (position estimation period: ΔT), and a phase voltage value creation unit 71, an induced voltage value calculation unit 72, an induced voltage The following operations are performed in the order of the amplitude value creating unit 73, the induced voltage reference value creating unit 74, the deviation creating unit 75, and the angular velocity correcting unit 76 to create the estimated rotor position θm and the estimated velocity ωm. Further, the current control unit 110 and the rotor position rotation number estimation unit 107 are operated in this order, and the position estimation cycle ΔT and the current control cycle are made the same.

  The position estimation period ΔT does not depend on the structure of the motor but depends on the processing capability of the microcomputer or DSP. For example, when the position estimation period ΔT is 67 μs, the number of magnetic poles of the rotor is 4, and the motor rotation speed is 1800 rpm (angular velocity 60π / second), the position estimation period ΔT is expressed by the electrical angle ΔΘ from the following equation. When expressed, it becomes 1.45 degrees. ΔΘ = 360 degrees × (4 poles / 2) × 67 μs × (1800 rpm / 60 s) = 1.45 degrees. Thus, ΔT is a very small value, and position estimation close to real time is performed. This response close to real time realizes high follow-up in sudden acceleration or rapid deceleration of the motor.

(Phase voltage value creation unit 71)
The phase voltage value creating unit 71 sets the voltage commands vu ^* , vv ^* , and vw ^* as the phase voltage values vu, vv, and vw as shown in equations (38), (39), and (40).

vu = vu ^* (38)
vv = vv ^* (39)
vw = vw ^* (40)

(Induced voltage value calculation unit 72)
The induced voltage value calculation unit 72 calculates an induced voltage calculated value (u-phase induced voltage calculated value eu, v-phase induced voltage) for each phase according to the expressions (1), (2), (3) or (4), (5), and (6). A calculated value ev and a w-phase induced voltage calculated value ew) are created.

(Induced voltage amplitude value creation unit 73)
The induced voltage amplitude value creation unit 73 first creates an induced voltage amplitude calculation value ec based on the result of adding the absolute values of the induced voltage values of the respective phases. As shown in the equation (8), the absolute value of the u-phase induced voltage value eu, the absolute value of the v-phase induced voltage value ev, and the absolute value of the w-phase induced voltage value ew are set to a certain value. A product of the coefficient KEC (see equation (9)) is used as an induced voltage amplitude calculation value ec. Further, a value obtained by applying a first-order digital low-pass filter (LPF) to the induced voltage amplitude calculation value ec is obtained from the equation (10) and is set as an induced voltage amplitude estimated value em.

(Induced voltage reference value creation unit 74)
The induced voltage reference value creating unit 74 obtains the induced voltage amplitude estimated value em created by the induced voltage amplitude value creating unit 73 and the estimated rotor position θm currently recognized by the control device 100, and calculates the induced voltage value. Induced voltage reference values eum, evm, ewm, which are reference values, are created.

(Deviation creation part 75)
The deviation creating unit 75 creates a deviation ε between the induced voltage calculation value esd and the induced voltage reference value esdm. First, the induced voltage calculated value es obtained by the induced voltage value calculating unit 72 and the induced voltage reference value esdm obtained by the induced voltage reference value creating unit 74 are expressed by the following equations (7) and (14): Using the current estimated rotor position θm, the coordinates are converted into DC components, respectively, and an induced voltage calculation value esd and an induced voltage reference value esdm are obtained. Next, as shown in equation (41), a value obtained by subtracting the induced voltage reference value edcm from the induced voltage calculation value edc of the dc axis is set as the deviation ε. Although the deviation on the dc axis is used here, the deviation on the qc axis may be used.

ε = edc-edcm (41)

(Angular speed correction unit 76)
The angular velocity correction unit 76 corrects the estimated rotor position θm so that the deviation ε converges to zero. Also, an estimated speed ωm is created. First, an advance amount θmp indicating how much the estimated rotor position θm is advanced every position estimation period ΔT is created. A value obtained by multiplying the deviation ε by the proportional gain κp and limiting the absolute value of the multiplication result so as not to exceed the proportional limit ζp as represented by the equation (42) is defined as a lead amount proportional term θmpp. Further, as shown in the equation (43), a value obtained by multiplying the deviation ε by the integral gain κi and limiting the absolute value of the multiplication result so as not to exceed the integral limit ζi is defined as an advance amount integral term θmpi. Then, as shown in the equation (44), the result of adding the advance amount integral term θmpi and the advance amount proportional term θmpp is defined as an advance amount θmp.

θmpp = κp · ε, −ζp ≦ θmpp ≦ ζp (42)
θmpi = κi · ε, −ζi ≦ θmpi ≦ ζi (43)
θmp = θmpp + Σθmpi (44)

  Next, the estimated rotor position θm is advanced by the advance amount θmp. The estimated rotor position θm is obtained by integrating the advance amount θmp as shown in equation (45). The new estimated rotor position θm is stored in the memory.

θm = Σθmp (45)

  The estimated speed ωm is obtained by applying a primary digital low-pass filter (LPF) to the advance amount θmp. Specifically, the estimated speed ωm is obtained based on the equation (46). Here, ωm (n) is the current estimated speed, and ωm (n−1) is the previous estimated speed. KTPW is a coefficient for changing the advance amount into a unit of speed. Furthermore, KLW is a coefficient of the low-pass filter and takes a value from 0 to 1, and the effect of the low-pass filter increases as the value decreases.

ωm (n) = KLW · (KTPW · θmp) + (1−KLW) · ωm (n−1) (46)

  In the present embodiment, the inverter output frequency ωm, which is an estimated speed, is adjusted based on the deviation depending on the position error Δθ obtained by the deviation creating unit 75. If the position error Δθ is positive during the forward rotation of the motor, it means that the estimated rotor position θm is behind the actual rotor position θ, and therefore the inverter frequency ωm is increased to accelerate the rotation of the control shaft. Conversely, if the position error Δθ is negative during forward rotation of the motor, it means that the estimated rotor position θm is ahead of the actual rotor position θ, so the inverter frequency ωm is lowered and the rotation of the control shaft is decelerated. Let

  Further, the angular velocity correction unit 76 may use means other than PI compensation having the same function in order to correct the estimated rotor position θm so that the deviation ε converges to zero.

(effect)
In the rotor position estimation method and the synchronous motor control device 100 of this embodiment, coordinate conversion is performed so that both the induced voltage reference value and the induced voltage calculation value are DC amounts. By treating both the induced voltage reference value and the induced voltage calculation value with a DC amount, the deviation ε depending on the phase error becomes a DC amount, and there is no fluctuation due to the AC amount, and the accuracy of the estimated rotor position θm and the estimated speed ωm is improved. . Moreover, the selection part (refer patent document 1 and FIG. 11) for selecting the phase which estimates is not required.

<< Second Embodiment >>
The second embodiment shown in FIG. 5 is substantially the same as the first embodiment shown in FIG. 1, but the configuration of the rotor position rotational speed estimation unit 207 is different from the first embodiment. In this embodiment, by using the condition that the induced voltage reference value edcm generated on the dc axis is 0, the calculation for obtaining the induced voltage reference value esdm is not performed.

(Rotor position rotational speed estimation unit 207)
FIG. 6 shows the configuration of the rotor position rotational speed estimation unit 207. The rotor position rotation speed estimation unit 207 is activated every set period (position estimation period: ΔT), and a phase voltage value generation unit 71, an induced voltage value calculation unit 72, a deviation generation unit 275, and an angular velocity correction unit. The following operations are performed in the order of 76 to create an estimated rotor position θm and an estimated speed ωm. Further, the current control unit 110 and the rotor position rotation number estimation unit 207 are operated in this order, and the position estimation cycle ΔT and the current control cycle are made the same.

(Phase voltage value creation unit 71, induced voltage value calculation unit 72)
The phase voltage value creation unit 71 and the induced voltage value calculation unit 72 perform the same operation as in the first embodiment.

(Deviation creation unit 275)
The deviation creating unit 275 creates a deviation ε between the induced voltage calculation value edc and the induced voltage reference value edcm on the dc axis. Here, utilizing the fact that the induced voltage reference value edcm is 0, the induced voltage reference value esdm is not calculated. Further, the induced voltage calculation value es obtained by the induced voltage value calculation unit 72 is coordinate-converted into a DC component using the current estimated rotor position θm, and the induced voltage calculation value esd is converted into the induced voltage calculation value esd as in the first embodiment. obtain. Here, paying attention to the induced voltage on the dc axis, as shown in the equation (47), the deviation ε becomes the induced voltage calculation value edc.

ε = edc (47)

(Angular speed correction unit 76)
Similar to the first embodiment, the angular velocity correction unit 76 corrects the estimated rotor position θm so that the deviation ε converges to zero. That is, in the present embodiment, this is equivalent to setting the deviation ε, that is, the induced voltage calculation value edc on the dc axis to zero. Further, the derivation of the estimated speed ωm and the estimated rotor position θm is the same as in the first embodiment.

(effect)
According to the present embodiment, since a selection unit and an LPF (low pass filter) for selecting a phase to be estimated are not required, the configuration can be simplified (the estimation algorithm can be simplified). Also, the amount of calculation is small compared to the first embodiment.

<< Third Embodiment >>
This embodiment shown in FIG. 7 differs from the first embodiment in the configuration of the rotor position rotation speed estimation unit 307. Specifically, in the first embodiment, in place of the U-phase current iu and the V-phase current iv, the dc-axis current idc and the qc-axis current iqc in the dc-qc coordinate axis are input to the rotor position rotational speed estimation unit 307. Different from form.

(Rotor position rotational speed estimation unit 307)
FIG. 8 shows the configuration of the rotor position rotational speed estimation unit 307. The rotor position rotation speed estimation unit 307 is activated every certain set cycle (position estimation cycle: ΔT), and includes a phase voltage value creation unit 71, a motor terminal voltage amplitude value creation unit 373, a voltage phase creation unit 377, and a motor. The terminal voltage reference value creation unit 374, the deviation creation unit 375, and the angular speed correction unit 76 are operated in the following order to create the estimated rotor position θm and the estimated speed ωm. Further, the current control unit 110 and the rotor position rotation number estimation unit 307 are operated in this order, and the position estimation cycle ΔT and the current control cycle are made the same.

(Phase voltage value creation unit 71)
The phase voltage value creation unit 71 sets the voltage commands vu ^* , vv ^* , and vw ^* as phase voltage values vu, vv, and vw based on the above-described equation (17).

(Motor terminal voltage amplitude value creation unit 373)
The motor terminal voltage amplitude value creation unit 373 first creates a motor terminal voltage amplitude calculation value vc based on the result of adding the absolute values of the motor terminal voltage values of each phase. As shown in the equation (22), there is a certain setting in the addition result of the absolute value of the u-phase motor terminal voltage value vu, the absolute value of the v-phase motor terminal voltage value vv, and the absolute value of the w-phase motor terminal voltage value vw. A value obtained by multiplying the obtained coefficient KEC (see equation (23)) is a motor terminal voltage amplitude calculation value vc. Further, a value obtained by applying a first-order digital low-pass filter (LPF) to the motor terminal voltage amplitude calculation value vc is obtained from the equation (10), and is set as a motor terminal voltage amplitude estimated value vm.

(Voltage phase creation unit 377)
The voltage phase creation unit 377 acquires idc, iqc created by the biaxial current conversion unit 106 and the current estimated speed ωm, and uses the equation (18) to calculate the dc axis voltage value vdc and the qc axis voltage value. Find vqc. Further, using equations (19) and (20), the phase βv from the qc axis where the induced voltage is generated to the δ axis corresponding to the motor terminal voltage position θv is obtained, and the motor terminal voltage position θv is obtained from the phase βv.

(Motor terminal voltage reference value creation unit 374)
The motor terminal voltage reference value creation unit 374 generates the motor terminal voltage amplitude estimated value vm created by the motor terminal voltage amplitude value creation unit 373, the estimated rotor position θm obtained by the angular velocity correction unit 76, and the voltage phase creation. The phase βv from the qc axis to the δ axis created by the unit 377 is acquired, and the motor terminal voltage reference values vum and vvm, which are reference values of the motor terminal voltage value, using the equations (25), (26), and (27). , Vwm.

(Deviation creation part 375)
The deviation creating unit 375 creates a deviation ε between the motor terminal voltage calculation value vsd and the motor terminal voltage reference value vsdm. First, the motor terminal voltage calculation value vs of each phase obtained by the phase voltage creation unit 71 and the motor terminal voltage reference value vsm of each phase obtained by the motor terminal voltage reference value creation unit 374 are expressed by (21) (28 As shown in the equation (2), the current motor terminal voltage position θv is used to perform coordinate conversion into a direct current component to obtain a motor terminal voltage calculation value vsd and a motor terminal voltage reference value vsdm on the γ-δ axis. Next, the deviation ε is obtained by subtracting the motor terminal voltage reference value vγm from the motor terminal voltage calculation value vγ of the γ-axis as shown in the equation (48). Although the deviation on the γ axis is used here, the deviation on the δ axis may be used.

ε = vγ−vγm (48)

(Angular speed correction unit 76)
Similar to the first embodiment, the angular velocity correction unit 78 corrects the estimated rotor position θm so that the deviation ε converges to zero. That is, in this embodiment, this is equivalent to setting the deviation εγ on the γ axis to zero. The estimation speed ωm and the estimation rotor position θm are derived in the same manner as in the first embodiment.

(effect)
According to this embodiment, since a selection unit and an LPF (low pass filter) for selecting a phase to be estimated are not required, the configuration can be simplified (simplification of the control program). Also, the amount of calculation is small compared to the first embodiment.

<< Fourth embodiment >>
The present embodiment shown in FIG. 9 is substantially the same as the third embodiment shown in FIG. 7, but the configuration of the rotor position rotational speed estimation unit 407 is different from the third embodiment. In the present embodiment, the calculation for obtaining the motor terminal voltage reference value vsdm is not performed by using the condition that the motor terminal voltage reference value vγm generated on the γ-axis is zero.

(Rotor position rotational speed estimation unit 407)
FIG. 10 shows the configuration of the rotor position rotational speed estimation unit 407. The rotor position rotational speed estimation unit 407 is activated every set period (position estimation period: ΔT), and the phase voltage value creation unit 71, the voltage phase creation unit 377, the deviation creation unit 475, and the angular velocity correction unit 76. The following operations are performed in this order to create the estimated rotor position θm and the estimated speed ωm. Further, the current control unit 110 and the rotor position rotation number estimation unit 407 are operated in this order, and the position estimation cycle ΔT and the current control cycle are made the same.

(Phase voltage value creation unit 71, voltage phase creation unit 377)
The phase voltage value creating unit 71 and the voltage phase creating unit 377 perform the same operation as in the third embodiment.

(Deviation creation part 475)
The deviation creating unit 475 creates a deviation ε between the motor terminal voltage calculation value vγ and the motor terminal voltage reference value vγm on the γ axis. Here, the motor terminal voltage reference value vsdm is not calculated using the fact that the value of the motor terminal voltage reference value vγm is 0. In addition, the motor terminal voltage calculation value vs of each phase obtained by the phase voltage value creation unit 71 is coordinate-converted into a DC component using the current motor terminal voltage position θv, and the motor terminal is the same as in the third embodiment. A voltage calculation value vγ is obtained. Here, paying attention to the motor terminal voltage on the γ-axis, the deviation ε becomes the motor terminal voltage calculation value vγ on the γ-axis as shown in the equation (49).

ε = vγ (49)

(Angular speed correction unit 76)
Similar to the third embodiment, the angular velocity correction unit 76 corrects the motor terminal voltage position θv so that the deviation ε converges to zero. In the present embodiment, this is equivalent to setting the deviation ε, that is, the motor terminal voltage value vγ on the γ axis to zero. Further, the derivation of the estimated speed ωm and the estimated rotor position θm is the same as in the third embodiment.

(effect)
According to this embodiment, since a selection unit and an LPF (low pass filter) for selecting a phase to be estimated are not required, the configuration can be simplified (simplification of the control program). Also, the amount of calculation is small compared to the third embodiment.

  As described above, according to the present invention, it is possible to perform motor control with high resolution and high accuracy by obtaining a deviation for each short position estimation cycle and sequentially correcting the estimated position. Moreover, the responsiveness to the speed change is also improved. Since it is difficult for the estimated rotor position θm and the estimated speed ωm to be affected by ripples such as ripple, even in the control of a motor that essentially has vibration caused by periodic load disturbance, such as a rotary compressor motor, Improvement in controllability can be expected.

  The present invention can also be applied to control of other synchronous motors, for example, control of SPMSM. Moreover, although the application object of this invention is not specifically limited, For example, this invention is useful for the control apparatus of the synchronous motor used for heat pump refrigeration apparatuses, such as an air conditioning apparatus and a water heater.

The block diagram of the control apparatus of the synchronous motor of 1st Embodiment by this invention. 1 is a configuration diagram of a drive unit according to a first embodiment of the present invention. Analysis model diagram showing relationship between rotor coordinates (dq coordinates) and estimated rotor coordinates (dc-qc coordinates) The block diagram of the rotor position rotation speed estimation part of 1st Embodiment by this invention. The block diagram of the control apparatus of the synchronous motor of 2nd Embodiment by this invention. The block diagram of the rotor position rotation speed estimation part of 2nd Embodiment by this invention. The block diagram of the control apparatus of the synchronous motor of 3rd Embodiment by this invention. The block diagram of the rotor position rotation speed estimation part of 3rd Embodiment by this invention. The block diagram of the control apparatus of the synchronous motor of 3rd Embodiment by this invention. The block diagram of the rotor position rotation speed estimation part of 3rd Embodiment by this invention. The figure which shows the estimation method described in patent document 1

Explanation of symbols

100, 200, 300, 400 Synchronous motor control device 102 Synchronous motor 104 Drive unit 105a, 105b Current sensor 106 Two-axis current conversion unit 107, 207, 307, 407 Rotor position rotational speed estimation unit 109 Voltage command calculation unit 110 Current Control unit 111 Current command generation unit 112 Rotational speed control units 113a to 113f Switching elements 114a to 114f Free-wheeling diode 116 Base driver 117 Smoothing capacitor 118 DC power supply

Claims (13)

Estimating the induced voltage of each phase of the synchronous motor based on the phase voltage value of the stator winding, the phase current value of the stator winding, the resistance of the stator winding, and the inductance of the stator winding When,
Creating an induced voltage calculation value obtained by coordinate-transforming the induced voltage of each phase to a control axis rotating at an arbitrary speed;
Creating an induced voltage reference value obtained by coordinate-converting the reference value of the induced voltage on the stationary coordinate to the control axis;
Creating a deviation between the induced voltage calculation value and the induced voltage reference value;
Estimating the rotor position by correcting the position of the control shaft so that the deviation converges to zero;
A rotor position estimation method for a synchronous motor.   The synchronous motor rotor position estimation method according to claim 1, wherein the coordinate of the control axis is an estimated coordinate recognized as a rotor position by a control device that performs control of the synchronous motor based on the method.   The synchronous motor rotation according to claim 1, wherein in the step of creating the induced voltage reference value, an amplitude estimation value indicating an amplitude of the induced voltage reference value is corrected based on the phase current value and the phase voltage value. Child position estimation method.   The synchronous motor rotor according to any one of claims 1 to 3, wherein in the step of estimating the rotor position, the position of the control shaft is corrected so that a magnetic flux direction component of the deviation converges to zero. Position estimation method.   In the step of estimating the induced voltage of each phase, assuming that the phase current flowing through the stator winding is a sine wave, the resistance of the stator winding and the stator from the phase voltage of the stator winding The method for estimating a rotor position of a synchronous motor according to any one of claims 1 to 4, wherein an induced voltage of each phase is estimated by reducing a voltage drop based on a winding inductance.   The synchronous motor rotor position estimation method according to claim 5, wherein the induced voltage of each phase is estimated based on a simplified phase voltage equation of the stator winding. A voltage command calculating unit that calculates and outputs a voltage command corresponding to a deviation between the current command and the detected current using an estimated position of the rotor so that a detected current follows the current command;
A current sensor for detecting a phase current value of a synchronous motor to which a three-phase AC voltage reflecting the voltage command is applied;
An inverter that converts a DC voltage into the three-phase AC voltage according to the voltage command and applies it to the synchronous motor;
Estimating the induced voltage of each phase of the synchronous motor based on the phase voltage value of the stator winding, the phase current value of the stator winding, the resistance of the stator winding, and the inductance of the stator winding Using the current estimated position of the rotor, an induced voltage calculation value obtained by converting the induced voltage of each phase into a DC component, and an induced voltage reference value on the estimated position, and the induced voltage calculated value and A rotor position estimating unit that creates a deviation from the induced voltage reference value and corrects the estimated position so that the deviation converges to zero, thereby estimating a rotor position;
A control device for a synchronous motor. Creating a motor terminal voltage calculation value obtained by coordinate-converting the phase voltage of each phase of the synchronous motor to a control axis rotating at an arbitrary speed;
Creating a motor terminal voltage reference value obtained by converting the reference value of the motor terminal voltage on the stationary coordinate to the control axis; and
Creating a deviation between the motor terminal voltage calculation value and the motor terminal voltage reference value;
Estimating the rotor position by correcting the position of the control shaft so that the deviation converges to zero;
A rotor position estimation method for a synchronous motor.   The coordinate of the control axis is the coordinate of the motor terminal voltage that rotates at a certain phase with the same angular velocity as the estimated coordinate at which the control device that executes control of the synchronous motor based on the method recognizes the position of the rotor. The method for estimating a rotor position of a synchronous motor according to claim 8.   The synchronous motor rotor according to claim 8 or 9, wherein in the step of creating the motor terminal voltage reference value, an amplitude estimation value indicating an amplitude of the motor terminal voltage reference value is corrected based on the phase voltage value. Position estimation method.   The synchronous motor rotor according to claim 8, wherein in the step of estimating the rotor position, the position of the control shaft is corrected so that a γ-axis component of the deviation converges to zero. Position estimation method.   In the step of creating the motor terminal voltage calculation value, in order to create a three-phase AC voltage to be applied to the synchronous motor by an inverter, a voltage command created corresponding to a deviation between a current command and a detected current is set to each phase. The rotor position estimation method for a synchronous motor according to any one of claims 8 to 11, which is used as a motor terminal voltage calculation value. A voltage command calculating unit that calculates and outputs a voltage command corresponding to a deviation between the current command and the detected current using an estimated position of the rotor so that a detected current follows the current command;
A current sensor for detecting a phase current value of a synchronous motor to which a three-phase AC voltage reflecting the voltage command is applied;
An inverter that converts a DC voltage into the three-phase AC voltage according to the voltage command and applies it to the synchronous motor;
A motor in which the phase voltage of each phase of the synchronous motor is coordinate-converted to a control axis that rotates at an arbitrary speed and a motor terminal voltage calculation value is created, and a motor terminal voltage reference value on a stationary coordinate is coordinate-converted to the control axis A terminal voltage reference value is created, a deviation between the motor terminal voltage calculation value and the motor terminal voltage reference value is created, and the estimated position is corrected so that the deviation converges to zero, thereby estimating the rotor position. A rotor position estimating unit,
A control device for a synchronous motor.

Images