JP2003209989A - Method and apparatus for detecting rotational position of brushless DC motor and refrigerator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect the rotational position of a rotor accurately regardless of the temperature variation of a motor. <P>SOLUTION: A current difference operating section 21 determines current differences ΔIdm and ΔIqm between estimated currents Idm and Iqm calculated at a motor model current operating section 20 based on the equivalent model formula of a motor and detected currents Id and Iq. An integration adjuster 24 integrates the current difference ΔIqm to determine a speed electromotive force em. A proportion adjuster 26 and an integration adjuster 27 perform the proportional operation and integration of the current difference ΔIdm, respectively. Based on proportion/integration results and the speed electromotive force em, an estimated speed electromotive force em and an estimated rotational position θm are obtained. Since the integration adjuster 27 is provided additionally, estimation accuracy is enhanced. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

The present invention relates to a d-axis / q-axis estimated current calculated based on an equivalent model formula and a detected d-axis / q.
The present invention relates to a rotational position detection method for a brushless DC motor that detects the rotational position of a rotor based on an axial current, a rotational position detection device, and a refrigerator using the same.

[0002]

Brushless DC motors (hereinafter referred to as motors) are widely used in industry because of their high efficiency and relatively easy maintenance.
In order to drive this motor, position information of the rotor is indispensable, and position detectors such as encoders, resolvers, and Hall sensors are used. However, when an encoder, a resolver or the like is attached to the rotary shaft, the motor becomes large and the cost increases. Further, when the rotor is exposed to a high temperature refrigerant like a compressor motor of a refrigerator, it becomes difficult to dispose the Hall sensor. In consideration of such circumstances, various methods for estimating the rotational position based on the motor current detected by the current sensor, that is, sensorless control methods have been proposed.

For example, IEEJ Transactions D, Vol. 117, No1,
In p98-104 (1997), "sensorless salient pole type brushless DC motor control based on speed electromotive force estimation" is shown.
This control method is an extension of the sensorless control method for a cylindrical motor based on speed electromotive force estimation to a salient pole motor. The estimation algorithm used here is a speed electromotive force based on the error current between the actual current and the estimated current based on the internal model equation, as shown in equations (11) and (12) of the above paper. And the rotational position is estimated.

However, this control method is good when the temperature change of the motor is small or when the magnet of the rotor having a small temperature dependence is used, but under other environments and conditions, the temperature change may be different. The magnetic flux of the magnet (that is, the speed electromotive force constant KE) and the winding resistance R change and deviate from the constants of the internal model formula, and an error occurs in the estimated rotational position. This error reduces the motor efficiency and causes an increase in vibration. For example, when the above control method is applied to drive a compressor motor of a refrigerator, there is a possibility that power consumption and noise increase.

In order to reduce the estimation error, it is possible to increase the estimated proportional gain of the rotational position due to the d-axis error current. However, if the estimated proportional gain is increased, the harmonic components appearing in the current are also increased, which causes a new problem that the torque ripple increases and an unstable drive state easily occurs.

The present invention has been made in view of the above circumstances, and an object thereof is a rotational position detecting method and a rotational position detecting method for a brushless DC motor capable of accurately detecting the rotational position of a rotor regardless of the temperature change of the motor. An object is to provide a device and a refrigerator using the device.

[0007]

In order to achieve the above object, a method for detecting a rotational position of a brushless DC motor according to a first aspect of the present invention detects a current flowing through the brushless DC motor, and detects the current by rotational coordinate conversion. Is separated into a d-axis current which is a field direction component of the brushless DC motor and a q-axis current which is a component orthogonal to the field direction component, and the d-axis which is a field direction component based on the equivalent model formula of the brushless DC motor. The estimated current and the q-axis estimated current that is a component orthogonal thereto are calculated, and the d-axis current deviation that is the difference between the d-axis estimated current and the d-axis current and the q-axis estimated current and the q-axis current are calculated. Of the d-axis current deviation is calculated by calculating the q-axis current deviation which is the difference between the two, and calculating the speed electromotive force by integrating the q-axis current deviation. And it obtains the rotational position of the rotor of the brushless DC motor based on the DoOkoshi power.

According to this method, the detected actual current flowing in the brushless DC motor is separated into the d-axis current and the q-axis current by the rotational coordinate conversion, and the preset equivalent model expression of the brushless DC motor is obtained. Based on, the d-axis estimated current and the q-axis estimated current are calculated. Then, based on these, the d-axis current deviation and the q-axis current deviation are calculated.

In general, since the speed electromotive force constant easily changes due to temperature change, the speed electromotive force obtained by integrating the q-axis current deviation is also easily affected by temperature. On the other hand, since the d-axis side does not include the element of speed electromotive force,
When the d-axis current deviation is constantly converged to 0 by the integral calculation, the rotational position of the rotor and thus the rotational speed can be obtained with little influence of temperature change. In this case, it is preferable to set the integral gain of the d-axis current deviation higher than the integral gain of the q-axis current deviation in order to make the integral calculation on the d-axis side dominant.

As a result, even if the magnetic flux of the motor magnet (that is, the speed electromotive force constant KE) fluctuates due to the temperature change and deviates from the constant of the equivalent model formula, the detected rotational position is caused by the temperature change. Error is hardly generated, and the rotational position can always be obtained with high accuracy. Then, if the brushless DC motor is driven using this rotational position detection method,
Motor efficiency can be improved and vibration can be reduced.

In this case, it is preferable to stop the integral calculation of the d-axis current deviation when the brushless DC motor is started (claim 2). Since the rate of change of the rotational speed and the load is large during the motor startup, that is, from the stopped state to the time when the predetermined rotational state is reached, high responsiveness is required for estimating the rotational position. Therefore, in this starting period, the rotational position is estimated and calculated only by the proportional calculation with respect to the d-axis current deviation, and then the integral calculation is started when the rotational speed and the load become stable. As a result, the error in the rotational position due to the temperature change of the motor can be reduced, and the stability at the time of starting is improved and the starting characteristics are improved.

Under a predetermined rotation condition, the speed electromotive force may be calculated using the target rotation speed instead of calculating the speed electromotive force by integrating the q-axis current deviation. . When the speed electromotive force of the motor contains a harmonic component, the harmonic component also appears in the speed electromotive force obtained by the integral calculation of the q-axis current deviation. Since this harmonic component enters the speed control loop, the gain of the speed control loop must be reduced so that the speed control loop does not respond to this harmonic component. However, when the gain is lowered, there are inconveniences such as a decrease in responsiveness and an increase in steady-state deviation. On the other hand, if the speed electromotive force is obtained from the target rotation speed, the entry of harmonic components into the speed control loop is reduced, so the gain of the speed control loop can be increased and high responsiveness and high stability are ensured. it can. Also, vibration of the motor is reduced.

According to the means described in claims 4 to 6, the rotational position detecting method described above can be realized as a rotational position detecting device. Then, a compressor provided in the refrigeration cycle and a brushless D that drives this compressor
The brushless DC motor is driven based on the C motor, the rotational position detecting device according to any one of claims 4 to 6, and the rotational position of the rotor of the brushless DC motor output from the rotational position detecting device. If a refrigerator is provided with the drive device that operates (Claim 7), the brushless DC motor exposed to the high-temperature refrigerant can be sensorlessly driven without using the rotational position detector. In this case, the rotational position detecting device can always obtain a highly accurate rotational position regardless of the temperature change of the motor, so that the efficiency of the refrigerator can be improved and the vibration caused by the motor can be reduced.

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION (First Embodiment) A first embodiment of the present invention will be described below with reference to FIGS. FIG. 2 is an electrical configuration diagram showing a brushless DC motor drive device. In FIG. 2, the drive device 1 includes a target torque calculation unit 2, a target current calculation unit 3, a target voltage calculation unit 4, a dq / αβ conversion unit 5, a PWM signal generation unit 6, an inverter circuit 7, and a motor current detection unit 8. (Corresponding to current detecting means), three-phase / two-phase converting section 9, αβ / dq converting section 10 (corresponding to coordinate converting means), and position estimating section 11. Further, the rotational position detecting device includes a motor current detecting unit 8, a three-phase / two-phase converting unit 9, and αβ / d among them.
The q conversion unit 10 and the position estimation unit 11 are included.

The inverter circuit 7 is constructed as a voltage type inverter in which switching elements such as IGBTs are connected in a three-phase bridge, and the motor current detecting section 8 has a-phase and b-phase.
Each phase motor current Ia, Ib (hereinafter simply referred to as current)
The hole CT for detecting Other components are software-processed by the control means including a processor such as DSP.

A three-phase brushless DC motor 12 (hereinafter referred to as a motor 12) driven by a drive unit 1 drives a reciprocating compressor 13 provided in a refrigerating cycle of a refrigerator, and has a rotor with four poles. It is a salient pole type motor having a permanent magnet. The motor 12 is the compressor 1
It is installed in the inside of the No. 3 and is placed in an environment where it is exposed to a high temperature refrigerant. Therefore, a hall sensor (position detector) for detecting the position of the rotor cannot be installed, and the drive device 1 is so-called sensorless drive.

Since the drive unit 1 performs so-called vector control for independently controlling the current in the field direction and the current in the direction orthogonal thereto, a speed control loop for controlling the rotation speed of the motor 12 and the motor 12 are provided. And a current control loop for controlling the current.

FIG. 3 shows the definition of coordinates used in the following description. That is, the d-axis is oriented in the direction of the magnetic flux (field) of the magnet of the rotor, and the q-axis is oriented 90 degrees ahead of the d-axis (that is, the direction of the speed electromotive force e). Further, the γ-axis is oriented in the direction of the field recognized by the control means of the driving device 1, and the δ-axis is oriented 90 ° ahead of the γ-axis (that is, the orientation of the estimated speed electromotive force em). When there is no error in the estimated calculated rotational position, the γ-axis and the δ-axis coincide with the d-axis and the q-axis, respectively. In the following description, for convenience, the d-axis and the q-axis are used to represent the state quantities related to the γ-axis and the δ-axis used inside the drive device 1.

First, the speed control loop will be described.
The target rotation speed ωRef of the motor 12 is input from an external control device, and the estimated rotation speed ω
m is output. The target torque calculation unit 2 uses the subtractor 14 to calculate the rotation speed deviation Δω according to the following equation (1).

[0020]

[Equation 1]

The target torque calculation unit 2 then uses the PID adjuster 1 to make the rotational speed deviation Δω converge to zero.
5 is used to calculate the target torque TrqRef shown in the following equation (2).

[0022]

[Equation 2]

The target current calculation unit 3 calculates the target current IdRef in the field direction (d-axis direction) necessary to output the target torque TrqRef and the quadrature with the target current TrqRef according to the following expressions (3) and (4). Target current IqRe in the direction (q-axis direction)
Find f and use it as the input to the current control loop.

[0024]

[Equation 3]

Next, the current control loop will be described.
The three-phase / two-phase converter 9 converts the three-phase currents Ia, Ib detected by the motor current detector 8 into two-phase currents Iα, Iβ according to the following equation (5).

[0026]

[Equation 4]

The αβ / dq conversion unit 10 performs rotational coordinate conversion (so-called dq) from the stator coordinates of the motor 12 to rotor coordinates.
Conversion), and the currents Iα, I according to the following equation (6)
β is converted into currents Id and Iq. Here, θm is the estimated rotational position of the rotor calculated by the position estimation unit 11.

[0028]

[Equation 5]

The target voltage calculation unit 4 uses the subtracters 16 and 17 in accordance with the following formulas (7) and (8) to obtain the currents Id, IdRef and IqRef from the target current calculation unit 3, respectively. Iq is subtracted to obtain current deviations ΔId, ΔI
Calculate q.

[0030]

[Equation 6]

Then, the target voltage calculation unit 4 sets P in order to converge the current deviations ΔId and ΔIq to 0, respectively.
Using the ID adjusters 18 and 19, the following expressions (9) and (10)
The voltages Vd and Vq shown in the formula are calculated.

[0032]

[Equation 7]

The dq / αβ conversion unit 5 uses the above-mentioned αβ / d.
Contrary to the q-converter 10, the rotor coordinate is converted into the stator coordinate by rotating coordinate conversion, and the voltage V is converted according to the following equation (11).
d and Vq are converted into voltages Vα and Vβ.

[0034]

[Equation 8]

The PWM signal generator 6 based on the space vector system converts the two-phase voltages Vα and Vβ into three-phase voltages Va and V.
While converting into b and Vc, a PWM signal given to the switching element forming the inverter circuit 7 is generated.
In this embodiment, the carrier wave of the PWM signal is 6 kHz.
(The cycle is set to 166 μsec), and the inverter circuit 7 outputs the pseudo sine wave voltage according to the PWM signal. By the above speed control loop and current control loop, the motor 12 (compressor 13) is driven to the target rotation speed ω.
It can be driven at a rotation speed that matches Ref.

Next, the electrical configuration of the position estimation unit 11 which is a characteristic part of the present invention will be described with reference to FIG. Generally, the voltage on the dq coordinate axes of the salient pole motor 12
The current equation is expressed by the following equations (12) and (13).

[0037]

[Equation 9]

The motor model current calculation unit 20 (corresponding to current estimation calculation means) has an equivalent model formula represented by formula (14) in which the voltage-current formula of formula (12) is represented by the discrete value system of the sampling period T. The estimated currents Idm and Iqm are calculated based on this equivalent model formula. This equivalent model formula uses a motor constant at a predetermined temperature, and is stored in advance in a storage means such as a semiconductor storage element such as an EEPROM. Further, in the present embodiment, the sampling period T is set equal to the above-mentioned PWM period (166 μsec).

[0039]

[Equation 10]

The current deviation calculator 21 (corresponding to the current deviation calculator) uses the subtracters 22 and 23 to obtain the following equation (15),
According to the equation (16), the estimated currents Idm (n) and Iqm (n) output from the motor model current calculation unit 20 are output from the αβ / dq conversion unit 10, respectively.
Iq (n) is subtracted to obtain a current deviation ΔIdm (n), ΔIqm
Calculate (n).

[0041]

[Equation 11]

Here, the estimated axes (γ axis, δ axis) shown in FIG.
If the angle error of (indicated by Δθ in FIG. 3) is small,
The current deviation ΔIdm (n) is approximately proportional to the angle error Δθ, and the current deviation ΔIqm (n) is approximately the estimated speed electromotive force e.
It is proportional to the error of m. Therefore, in the present embodiment, the estimated rotational speed ωm is calculated by feedback control so that the current deviation ΔIdm (n) converges to zero. The method will be described below.

First, the q-axis integral adjuster 24 (corresponding to speed electromotive force computing means) converges the current deviation ΔIqm (n) to 0 according to the following equation (17), and
The estimated speed electromotive force em (n) used in the equation 4) is calculated.

[0044]

[Equation 12]

The estimated speed electromotive force em (n) calculated here
Is divided by the speed electromotive force constant KE in the divider 25 and becomes the basic term of the estimated rotation speed ωm ((18) below).
See formula). However, when the actual motor temperature is different from the reference temperature of the motor constant used in the equivalent model formula, this basic term has an estimation error due to the temperature. a d-axis proportional controller 26, an integral controller 27 and an adder 28,
29 is the current deviation ΔIdm (n) according to the following equation (18).
The estimated rotation speed ωm (n) is calculated by converging to 0.

[0046]

[Equation 13]

Although there are integral terms in the calculations of these equations (17) and (18), the integral gain KiEm
And Kθi are set so that the following equation (19) is satisfied, the integral term of the d-axis is dominantly exerted, and the estimated responsiveness of the estimated rotation speed ωm (n) is estimated.
It is higher than the estimated response of (n).

[0048]

[Equation 14]

The integrator 30 integrates the estimated rotation speed ωm (n) according to the following equation (20), and calculates the estimated rotation position θ of the rotor.
Find m (n). That is, in the present embodiment, the above (1
Although the estimated rotation speed ωm is obtained by the equation 8), the equation (18) above essentially obtains the estimated rotation position θm. The proportional adjuster 26, the integral adjuster 27, the adders 28 and 29, and the integrator 30 described above correspond to the rotational position calculating means in the present invention.

[0050]

[Equation 15]

Among the motor constants, the parameters having temperature dependence are the winding resistance R and the speed electromotive force constant KE. In the case of a general motor, the speed electromotive force value em is much more dominant than the impedance drop value R · Iq in the voltage / current equation shown in equation (12). In the present embodiment, as shown in the above equations (17) to (19), the current deviation Δ on the d-axis side that is not affected by the speed electromotive force constant KE.
Idm (n) is set to 0 by using the high gain integral regulator 27.
Has converged to. Therefore, the estimation error of the estimated rotational position θm (n) due to the temperature change can be significantly reduced as compared with the conventional configuration in which only the proportional adjuster 26 is used to estimate the rotational position (rotational speed).

The results of the test conducted to verify the effect of the above-described rotational position detecting method will be described below. In the test, an encoder (360 pulses / revolution) was attached to the salient pole type test motor 2, and the estimated rotation output from the position estimation unit 11 was changed while heating the motor 2 with a heater to change the motor temperature by 10 ° C. The difference (estimation error) between the position θm and the actual rotational position θ obtained based on the encoder signal was measured. The encoder was accurately aligned according to the coordinate definition shown in FIG. The rotation speed ω of the motor 2 is 30r
The ps and the torque are 0.2 Nm.

The encoder signal is input to the processor, and the rotational position θ based on the encoder signal is scaled to the same standard value as the estimated rotational position θm calculated inside the processor, and then the estimated rotational position θm and D /
It is output from the processor via the A converter. Due to the test equipment, the output cycle from the D / A converter is PW.
Since it is set equal to the period of the M signal, the maximum is 166.
An output delay of μsec occurs, but the rotation speed ω of the motor 2
Since it is low, there is no particular problem.

FIG. 4 shows the measurement results. The horizontal axis indicates the motor temperature (° C), and the vertical axis indicates the estimation error (%). In FIG. 4, the measurement points plotted by ● are those by the driving device 1 of the present embodiment, and the measurement points plotted by □ are “speed electromotive force” listed in “Problems to be solved by the invention”. This is based on "sensorless salient pole brushless DC motor control based on estimation" (hereinafter referred to as conventional control).

In the temperature range from −10 ° C. to 130 ° C. including the motor temperature range expected for application to the compressor, the conventional control causes an estimation error of about ± 20%, whereas the drive device of the present embodiment. 1 has an estimated error of 1%
It can be seen that the estimated error becomes substantially constant, and the estimation error of the estimated rotational position θm is greatly improved. This is considered to be the effect of adding the integral adjuster 27 described above.

As described above, the driving apparatus 1 of this embodiment has the equivalent model formula of the motor 12, and the estimated currents Idm and Iqm calculated based on this equivalent model formula,
The current deviations ΔIqm and ΔIqm from the currents Id and Iq detected by the motor current detection unit 8 are calculated to obtain the current deviation ΔIqm.
Is converged to 0 to calculate the estimated speed electromotive force em, and the current deviation ΔIdm is converged to 0 to calculate the estimated rotational speed ωm (that is, the estimated rotational position θm).

Further, since the integral adjuster 27 is provided on the d-axis side which does not include the element of the speed electromotive force e which easily changes due to the temperature change, and the current deviation ΔIdm is constantly converged to 0, The estimated rotational speed ωm and thus the estimated rotational position θm can be obtained with little influence of the change. Further, the integral gain KiEm for the current deviation ΔIdm is set to the integral gain Kθ for the current deviation ΔIqm.
Since it is set larger than i, the correction on the d-axis side has a strong effect, and the estimation error due to temperature change can be further reduced.

As a result, as verified by the test, the magnetic flux (velocity electromotive force constant KE) of the motor magnet changes due to the temperature change.
Even if the value fluctuates and deviates from the constant of the equivalent model formula, there is almost no error in the estimated rotational position θm, and high position accuracy can be obtained. As a result, motor efficiency can be improved and vibration can be reduced.

Further, the motor 12 for driving the compressor 13 of the refrigerator is exposed to the high temperature refrigerant during the operation of the compressor 13. However, if the motor 12 is controlled by the sensorless vector using the driving device 1, the temperature of the motor 12 changes. Independently, the motor efficiency and thus the refrigerator efficiency can be improved, and the vibration caused by the motor 12 can be reduced.

(Second Embodiment) Next, a second embodiment in which the starting characteristic is improved as compared with the first embodiment will be described with reference to FIG.
Will be described with reference to. FIG. 5 shows the electrical configuration of the position estimation unit as in FIG. The position estimation unit 31 shown here is different from the position estimation unit 11 shown in FIG. 1 in that an opening / closing means 32 is provided between the integration adjuster 27 and the adder 28. Other components in FIG. 5 are the same as those in the first embodiment, and the overall configuration of the driving device is also the same as in FIG.
As shown in. Also in this embodiment, each processing shown in the position estimation unit 31 and the like is software-processed by the processor.

To start the motor 12, first the currents Id and Iq
After positioning the rotor for about 1.5 seconds with the constant values of 0A and 0A respectively, the currents Id and Iq are set to 0, respectively.
A. A starting torque is generated by switching to a constant value. Almost at the same time as the starting torque generation processing, the position estimation processing and the speed control processing are sequentially started. The magnitude of the current Id at the time of positioning, the magnitude of the current Iq at the time of starting torque generation, and whether to use the step response or the ramp response may be appropriately switched depending on the situation.

In any case, however, it is difficult to estimate the rotational position at the time of starting, and if the rotational position is greatly deviated due to the estimation delay, the generated torque may become negative and the motor 12 may rotate in the reverse direction. Therefore, a higher responsiveness than a high accuracy is required for estimating the rotational position immediately after starting. Therefore, in the present embodiment, the opening / closing means 32 is opened immediately after the start, and the estimated rotation speed ωm (n) is calculated by the following equation (21).

[0063]

[Equation 16]

After the start, when the conditions of the following equations (22) and (23) are both satisfied, the opening / closing means 32 is switched to the closed state, and thereafter, the equation (18) is the same as in the first embodiment. The estimated rotation speed ωm (n) is calculated in accordance with

[0065]

[Equation 17]

According to the starting control described above, the integral adjusting device 27 is disconnected during the starting period and the estimated rotational position θm is calculated only by the proportional adjusting device 26. Therefore, the estimated delay of the estimated rotational position θm becomes small and stable. Starting characteristics are obtained. Further, after the start-up period has elapsed, the operation of the integral adjuster 27 enables high-efficiency and stable driving as in the first embodiment. It should be noted that although the integration adjuster 27 does not operate during the start-up period, some positional deviation occurs, but since the start-up period is very short, there is almost no effect due to a decrease in motor efficiency, and an increase in power consumption when applied to a refrigerator also occurs. Absent.

(Third Embodiment) Next, a third embodiment in which the stability during steady driving is improved as compared with the first embodiment will be described with reference to FIG. FIG. 6 shows the electrical configuration of the position estimation unit as in FIG. The position estimation unit 33 shown here is different from the position estimation unit 11 shown in FIG.
The rotation speed output from the divider 25 and the target rotation speed ωRe
The difference is that a switching means 34 for selecting either one of f 1 and supplying it to the adder 29 is provided. The other components in FIG. 6 are the same as those in the first embodiment, and the overall configuration of the drive device is also as shown in FIG. Also in this embodiment, each process shown in the position estimation unit 33 and the like is software-processed by the processor.

In the first embodiment, the approximate rotation speed is calculated by em / KE (see the equation (18)). However, when the speed electromotive force e of the motor 12 includes a harmonic component, the estimated speed is calculated. A harmonic component is also superimposed on the electromotive force em.
In this case, the speed control loop responds to this harmonic component and vibration is likely to occur. Therefore, at low rotational speeds, the switching means 34 selects the rotational speed from the divider 25 to obtain the estimated rotational speed ωm according to the equation (18), while at higher rotational speeds, the switching means 34.
The target rotation speed ωRef is selected by and the estimated rotation speed ωm (n) is calculated by the following equation (24). The switching threshold is, for example, 40 rps.

[0069]

[Equation 18]

According to the drive control described above, the entry of harmonic components into the speed control loop can be reduced without lowering the proportional gain Kθp and the integral gain Kθi which are the estimated gains, so that the gain of the speed control loop is increased. Therefore, high responsiveness and high stability can be secured, and vibration of the motor 12 can be reduced.

(Other Embodiments) The present invention is not limited to the above-described embodiments, and the following modifications or expansions are possible. The motor 12 is not limited to the salient pole type, but may be a cylindrical type. Further, the brushless DC motor according to the present invention uses a permanent magnet for the field,
A motor (for example, a synchronous motor) whose rotation principle is an interaction between the magnetic flux and the magnetic flux generated by the armature current is widely included. The drive device 1 including the rotational position detecting device of the present invention is not limited to being applied to a compressor motor of a refrigerator, but can be widely applied to a compressor motor of an air conditioner and other brushless DC motors. A configuration may be used in which both the start control described in the second embodiment and the drive control described in the third embodiment are used.

[0072]

As is apparent from the above description, the method for detecting the rotational position of the brushless DC motor according to the present invention detects the d-axis current and the q-axis current and calculates the d-axis estimated current based on the equivalent model formula. The q-axis estimated current is calculated and the q-axis current deviation is integrated to obtain the speed electromotive force, and the rotation of the rotor is calculated based on the speed electromotive force and the result of the proportional calculation and the integral calculation of the d-axis current deviation. Find the position. As a result, the current deviation can be steadily converged to 0 on the d-axis side that does not include the speed electromotive force element by the integral calculation, so that a highly accurate rotational position can be obtained with almost no influence of temperature change. be able to. By driving the brushless DC motor using this rotational position detecting method, the motor efficiency can be improved and the vibration can be reduced.

[Brief description of drawings]

FIG. 1 is an electrical configuration diagram of a position estimation unit showing a first embodiment of the present invention.

FIG. 2 is an electrical configuration diagram showing a brushless DC motor driving device.

FIG. 3 is a diagram showing the definition of coordinates.

FIG. 4 is a diagram showing a test result of an estimation error of a rotational position with respect to a motor temperature.

FIG. 5 is a view corresponding to FIG. 1 showing a second embodiment of the present invention.

FIG. 6 is a view corresponding to FIG. 1 showing a third embodiment of the present invention.

[Explanation of symbols]

1 is a drive device, 8 is a motor current detection unit (current detection unit), 10 is an αβ / dq conversion unit (coordinate conversion unit), 12
Is a brushless DC motor, 13 is a compressor, 20 is a motor model current calculator (current estimation calculator), 21 is a current deviation calculator (current deviation calculator), 24 is an integral controller (speed electromotive force calculator), 34 Is a switching means.

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5H560 AA02 BB04 BB12 DA14 DB20                       EB01 RR03 SS07 TT07 TT15                       XA02 XA04 XA12 XA13                 5H576 AA12 BB02 BB04 CC05 DD02                       DD07 EE01 EE11 FF01 FF02                       GG02 GG04 HB01 JJ03 JJ22                       JJ23 JJ24 LL14 LL22 LL24                       LL41

Claims (7)

[Claims] 1. A current flowing through a brushless DC motor is detected, and the detected current is converted by rotational coordinate conversion into a d-axis current which is a field direction component of the brushless DC motor and a q-axis current which is a component orthogonal thereto. The d-axis estimated current, which is a field direction component, and the q-axis estimated current, which is a component orthogonal to the field direction component, based on the equivalent model formula of the brushless DC motor. The d-axis current deviation that is the difference from the axis current and the q-axis current deviation that is the difference between the q-axis estimated current and the q-axis current are calculated, and the speed electromotive force is calculated by integrating the q-axis current deviation. The brushless D is obtained based on the result of the proportional calculation and the integral calculation of the d-axis current deviation and the calculated speed electromotive force.
A method for detecting a rotational position of a brushless DC motor, characterized in that a rotational position of a rotor of a C motor is obtained. 2. The method for detecting the rotational position of a brushless DC motor according to claim 1, wherein the integral calculation of the d-axis current deviation is stopped when the brushless DC motor is started. 3. Under a predetermined rotation condition, the speed electromotive force is calculated using a target rotation speed instead of calculating the speed electromotive force by integrating the q-axis current deviation. Item 3. A method for detecting a rotational position of a brushless DC motor according to Item 1 or 2. 4. A current detecting means for detecting a current flowing through a brushless DC motor, and a current detected by the current detecting means is a d-axis current which is a field direction component of the brushless DC motor and a component orthogonal thereto. A coordinate conversion unit that performs rotational coordinate conversion so as to be separated into a certain q-axis current, a d-axis estimated current that is a field direction component based on the equivalent model formula of the brushless DC motor, and a component orthogonal to this. A current estimation calculation means for calculating the q-axis estimated current, and d output from the d-axis estimated current and the coordinate conversion means.
A d-axis current deviation that is a difference from the axis current and a current deviation calculation means that calculates a q-axis current deviation that is a difference between the q-axis estimated current and the q-axis current output from the coordinate conversion means; And a speed electromotive force calculating means for calculating a speed electromotive force by integrating the q-axis current deviation, a proportional means and an integrating means, and a result of the proportional calculation and integral calculation of the d-axis current deviation and the speed. Based on the speed electromotive force output from the electromotive force calculation means, the brushless D
A rotation position detecting device for a brushless DC motor, comprising: a rotation position calculating means for obtaining a rotation position of a rotor of a C motor. 5. The rotation position calculation means is configured to stop the integration calculation of the d-axis current deviation by the integration means when the brushless DC motor is started. Brushless DC motor rotation position detector. 6. The speed electromotive force computing means selectively outputs either one of a speed electromotive force obtained by integrating the q-axis current deviation and a speed electromotive force obtained by a target rotation speed. 6. The rotation position detecting device for a brushless DC motor according to claim 4 or 5, further comprising switching means for switching. 7. A compressor provided in a refrigeration cycle, a brushless DC motor for driving the compressor, a rotational position detecting device according to claim 4, and an output from the rotational position detecting device. The brushless D
A refrigerator configured to drive the brushless DC motor based on a rotational position of a rotor of the C motor.