JP2003164190A - Motor drive device and motor rotor position detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a motor drive device and a rotor position detection method, which detects rotor position information used for motor drive in favorable accuracy without using a revolution position sensor. <P>SOLUTION: The motor drive device has a PWM modulation circuit 31 which determines a normal PWM pulse signal, by converting an output voltage value to a motor 2 which is derived with predetermined calculation to a duty value, by quantizing the duty value with a resolution of one sixth period of a carrier cycle, by determining the duty value of the PWM pulse signal upon referring to a table including PWM pattern which is determined so that ascending edges and descending ones of pulses do not overlap between respective phases and only one pulse is included in one carrier cycle, and by adding a portion omitted at the time quantized to the duty value and a position detection circuit 30 to calculate a present revolution phase from a predetermined calculation formula based on a present variation quantity. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a drive device for a motor, particularly a synchronous motor having a permanent magnet rotor, and a method for detecting the rotor position of the motor used in drive control thereof.

[0002]

2. Description of the Related Art In drive control for driving a compressor of a refrigerating and air-conditioning apparatus at a variable speed by using a single-phase AC power source, the position of a rotor of a motor is detected, and an energized phase is detected according to the rotor position. Switch.

The rotor position is usually detected by using a position sensor, but conventionally, there is a variable speed driving method for a motor without using the rotor position sensor. For example, for a three-phase electric terminal of a motor, a constant voltage is applied to two of the phases, the induced voltage is observed from the voltage in the remaining non-energized phase, the rotational phase is determined, and the phase is energized. There is a method to switch to cyclic.

Further, a commercial single-phase AC power supply is used as a basic power supply for driving, AC is rectified and smoothed to a nearly DC state, and then a pseudo switching AC is applied by pulse width modulation using a semiconductor switching device. There is a method of driving the motor as it is. Further, as a method of driving the motor with higher efficiency, there is a method of driving with a sine wave current.

FIG. 15 is a diagram showing a circuit configuration of a drive device for driving a motor by a general sine wave drive.
In sine wave drive, current always flows through the motor,
Since the method of obtaining the induced voltage information from the non-energized terminal voltage cannot be used, the phase current of the motor is detected by the current sensors 5u, 5v, 5w
In addition, the terminal voltage of the smoothing capacitor 16 is detected, the applied voltage of the motor is calculated using the pulse width modulation value, the induced voltage information of the motor is calculated from the motor parameter, and the rotation phase information is calculated. To do. In addition, since the sum of three current values of the current sensor is always 0, one may be omitted and two current values may be used.

As another method, there is also a method of detecting the rotational phase by utilizing the characteristic that the inductance of each phase of the motor winding changes depending on the rotational phase. FIG. 16 is a diagram showing the inductance characteristic and the induced voltage characteristic of the motor. As shown in the figure, the inductance of the motor has a characteristic of changing twice during one cycle of the induced voltage. As a position detection method using such characteristics,
Takeshita et al., "Sensorless salient pole type brushless D
Starting Method and Stability of C Motor "(The Institute of Electrical Engineers of Japan, SPC-
95-95). in this way,
A pulse voltage is separately applied to the motor to detect the position, and the inductance value is checked from the current value at that time.

Further, "Position sensorless IPM motor drive system using position estimation method based on saliency" by Ogasawara et al. (The Institute of Electrical Engineers of Japan, Applied Applications, Vol. 5, No. 5, 199)
In the method disclosed in 8), in order to reliably obtain the current value for obtaining the inductance value, the waveform of the PWM pulse in one carrier cycle is detected so that the change edge of the voltage of each phase can be detected reliably. I am devising. That is, as shown in FIG. 17, the PWM pulse for driving the inverter is devised so that the rising edge and the falling edge of the PWM pulse appear twice each in one carrier cycle. By driving the inverter using the PWM pulse with such a waveform, the change in the voltage edge of each phase of the motor winding can be reliably detected, and by checking the current corresponding to the voltage change edge, Position detection is possible without superimposing a pulse voltage on.

[0008]

However, in the position detecting method using the PWM pattern shown in FIG. 17, a normal PW is used.
The number of edges per carrier period is doubled as compared with M modulation. Therefore, the three-phase PWM inverter circuit 4
On the other hand, switching loss increases, and high-speed response is required. Therefore, there is a need for an algorithm and a circuit that implements it, which enables stable and high-speed position detection without generating extra noise even when the advance angle is large.

Further, generally, in the motor drive device shown in FIG. 15, a smoothing capacitor 16 having a large capacity is provided to reduce voltage pulsation. However, the smoothing capacitor 16 having a large capacity has an adverse effect of increasing the size of the device. is there. Although the size of the smoothing capacitor 16 can be reduced by reducing the capacity of the smoothing capacitor 16, the power supply voltage fluctuates due to a momentary power failure and the DC voltage Vdc is 0V.
When it drops to the vicinity and becomes lower than the motor terminal voltage,
Since the motor is braked, it is necessary to pass a current that weakens the magnetic flux. When the current that weakens the magnetic flux is increased, so-called advance control is required to advance the current that generates torque with respect to the induced voltage. However, in the rotational position detection method using the induced voltage shown in FIG. 15, the position of the induced voltage is increased. It is difficult to determine the rotational phase because the point where the change with respect to the shift is small is used.

The present invention has been made to solve the above problems, and is capable of accurately detecting rotor position information used for variable speed drive of a motor without using a rotational position sensor and motor rotor position detection. The purpose is to provide a method.

[0011]

A first motor driving device according to the present invention uses a rectifying means for rectifying a single-phase AC voltage and a pulse width modulation (PWM) of the rectified DC voltage using a switching element. Inverter means for converting into a pseudo three-phase AC voltage of arbitrary frequency and voltage by carrying out,
It is a motor drive device that drives a motor with a pseudo three-phase AC voltage. The motor drive device, means for obtaining an output voltage value to be output to each phase of the motor winding by a predetermined calculation,
A table in which a duty value and a PWM output pattern to be output for the duty value are associated with each other, and rising and falling edges of pulses do not overlap between phases,
In addition, a table including a plurality of PWM output patterns determined so that only one pulse is included in one PWM carrier cycle, an output voltage value is converted into a duty value in pulse width modulation, and the obtained duty value is PW.
A PWM pulse signal to be quantized with a resolution of a period that is an integral multiple of 1/6 of the M carrier cycle, select a PWM output pattern by referring to the table based on the quantized duty value, and output from the PWM output pattern. The duty value and phase of the normal PWM pulse signal for driving the switching element of the inverter means are added to the determined duty value by adding the cutoff amount at the time of quantization to the duty value and phase. The pulse modulation means for determining and the current change of each phase of the motor winding are detected, and based on the current change rate δI'u, δI'v, δI'w which is the current change amount per unit time and unit voltage change. And a position detecting means for calculating the current rotation phase θ ^{^} by using a predetermined calculation formula. This allows 1 for any phase, without position sensor
Position detection becomes possible within the carrier cycle.

The current rotational phase θ ^{^} can be calculated based on the current change rates δI'u, δI'v, δI'w by the following formula.

[Equation 3]

Also, before starting the motor, the motor specified 2
The motor may be positioned by passing a constant current between the two phases, and the phase at that time may be set as the initial value of the rotor phase. As a result, the N of the permanent magnet included in the rotor is
The pole and the S pole can be discriminated, and the subsequent position detection can be accurately performed.

The pulse modulating means is a counter for counting in synchronization with a predetermined clock, and an initial value for storing a set value for determining the value of the output voltage for each switching element of the inverter means when a PWM pulse is generated. A register, an initial value buffer register that stores data to be transferred to the initial value register at each carrier cycle, and a first register that stores a counter value that raises the output voltage value for each switching element from Low to High. , A first buffer register that stores data to be transferred to the first register at each carrier cycle, and a second register that stores a counter value that lowers the output voltage value for each switching element from High to Low. , A second buffer register for storing data to be transferred to the second register at each carrier cycle, It may comprise means for storing the current value of the motor windings at a timing of the value of the output voltage for each switching element is changed. This makes it easy to independently set the PWM pulse for each phase of the motor. The clock may be initialized every carrier cycle.

In the second drive device according to the present invention, the phase range is preassigned every 60 degrees according to the magnitude relationship between the voltages of the respective phases of the motor winding, and the input voltage of the single-phase AC is detected. Means, and when detecting that the input voltage has dropped below a predetermined voltage, detects the terminal voltage of each phase of the motor, based on the magnitude relationship between the detected voltage of each phase of the motor winding,
Means for determining the rotor phase to a value within the preassigned phase range. As a result, when a short-term abnormality occurs in the power supply, the rough rotation phase can be captured without stopping the power supply to the motor and braking, and the input voltage returns to the normal state. When you do, you can continue the rotation immediately and smoothly.

A first position detecting method according to the present invention rectifies a single-phase AC voltage and performs pulse width modulation (PWM) on the rectified DC voltage by using a switching element to obtain an arbitrary frequency and voltage. This is a detection method of detecting a rotor position of a motor in motor drive control for converting the pseudo three-phase AC voltage and driving the motor using the pseudo three-phase AC voltage. The method is to obtain an output voltage value to be output to each phase of the motor winding by a predetermined calculation, convert the output voltage value to a duty value in pulse width modulation, and obtain the obtained duty value by 1/6 of the PWM carrier cycle. Is a table in which quantization is performed with a resolution of an integer multiple of the period, and a duty value and a PWM output pattern to be output for the duty value are associated with each other, and rising and falling edges of pulses do not overlap between phases, and , A PWM output pattern is selected based on the quantized duty value by referring to a table including a plurality of PWM output patterns determined so that only one pulse is included in one PWM carrier cycle, and the PWM output pattern is selected. Determine the duty value and phase of the PWM pulse signal that should be output from, and switch to the determined duty value during quantization. Terra The amount by adding, to determine the duty value and the phase of the normal PWM pulse signals,
Using the determined regular PWM pulse signal, the switching element is driven, the current change of each phase of the motor winding is detected, and the current change rate δI ′ which is the current change amount per unit time and unit voltage change. Based on u, δI'v, δI'w, the current rotation phase θ ^{^} is calculated using a predetermined calculation formula.

A second position detecting method according to the present invention detects a single-phase AC input voltage by pre-allocating a phase range every 60 degrees according to the magnitude relationship between the voltages of the phases of the motor winding. However, when it is detected that the input voltage has dropped below a predetermined voltage, the terminal voltage of each phase of the motor is detected, and based on the magnitude relationship between the detected voltage of each phase of the motor winding, Determine the phase of the rotor to a value within the assigned phase range.

[0018]

DETAILED DESCRIPTION OF THE INVENTION A motor drive device and method according to the present invention will be disclosed below with reference to the accompanying drawings. In the following description, in the motor current control, a rotating coordinate axis composed of a d-axis and a q-axis is used. The q-axis is a rotating coordinate axis of a current that can generate torque by a permanent magnet, and advances by 90 ° with respect to it. The rotation coordinate axis is the d axis.

(Embodiment 1) FIG. 1 is a block diagram showing the overall configuration of a motor drive device according to the present invention. The motor drive device converts the voltage input from the single-phase AC power supply 1 into a three-phase AC voltage to drive the motor 2 at a variable speed. The motor drive device is a rectifier circuit 3 composed of a diode bridge,
A choke coil 18 for improving the power factor, a small (small capacity) smoothing capacitor 16, a three-phase PWM inverter circuit 4 including a plurality of switching elements, and an on / off operation of the switching elements of the three-phase PWM inverter circuit 4. And a control circuit 7 for controlling the. Further, the motor drive device has current sensors 5u, 5v, 5w for detecting the currents of the windings of the respective phases of the motor 2. The choke coil 18
In the present invention, since the smoothing capacitor capacity can be reduced, it can be omitted when the demand for the power source power factor is not so high as compared with the conventional case where the capacitor capacity is large.

In the driving device, electric power is input from the single-phase AC power supply 1, passes through the rectifier circuit 3, and passes through the small smoothing capacitor 16 that removes the influence of the PWM pulse in the subsequent stage, and then through the three-phase PWM inverter circuit 4. And is sent to the motor 2. The control circuit 7 refers to the phase currents of the three phases of the windings of the motor 2 detected by the current sensors 5u, 5v, 5w, detects the DC voltage Vdc obtained by the smoothing capacitor 16, and sets the set rotation speed. Judging the situation with respect to (command value) ω *, pulse width modulation (PW
M) Output the command. The current sensor for detecting the current of the motor 2 does not necessarily have to be provided for each phase, and only the current sensors for two phases may be provided. This is because the sum of the three current values is always 0, and if the currents of the two phases are known, the current value of the remaining one phase can be calculated.

FIG. 2 shows a more detailed structure of the control circuit 7. The control circuit 7 includes a torque control block 20, coordinate conversion circuits 21 and 22, a compensation circuit 23, a position detection circuit 30, and
A PWM modulation circuit 31 is included.

In the control circuit 7, the position detection circuit 30 calculates the motor 2 rotation phase information θ and the rotation speed information ω.
Details of the operation of the position detection circuit 30 will be described later. The rotation phase information θ is sent to the coordinate conversion circuits 21 and 22. Further, the rotation speed information ω is sent to the torque control block 20. The torque control block 20 uses the rotational speed information ω obtained from the position detection circuit 30, the intermediate DC voltage Vdc, the set rotational speed (rotational speed command) ω *, and the parameters (resistance, inductance, etc.) of the motor 2 for d-axis. The set values Id * and Iq * of the current and the q-axis current are calculated. Current sensor 5u, 5v, 5w
The motor phase current information detected at is sent to the coordinate conversion circuit 21 and the current rotation phase information θ is used to rotate the rotation coordinate axis d.
It is converted into current values Id and Iq on the axis and the q axis. Converted
The d-axis current Id and the q-axis current Iq are sent to the comparison circuits 24 and 25 and compared with the d-axis current set value Id * and the q-axis current set value Iq * to obtain current error information for each current. The current error information is sent to the compensating circuit 23 for improving the control characteristic, and converted into voltage set values Vd * and Vq * on the d-axis and the q-axis. The voltage setting values Vd * and Vq * are sent to the coordinate conversion circuit 22, and the voltage values Vu on the U, V, and W axes are calculated using the rotation phase information θ.
It is converted into *, Vv *, Vw * and sent to the three-phase PWM modulation circuit 31. The three-phase PWM modulation circuit 31 includes the voltage conversion circuit 22.
The PWM pattern is determined based on the voltage values Vu *, Vv *, Vw * from That is, in the three-phase PWM modulation circuit 31, the voltage value Vu from the voltage conversion circuit 22 follows the predetermined rule.
The duty value and phase of the PWM pulse signal are determined from *, Vv *, and Vw *. The detailed operation of the three-phase PWM modulation circuit 31 will be described later. In the three-phase PWM inverter circuit 4, the ON / OFF operation of the switching element is controlled by the PWM pulse signal from the three-phase PWM modulation circuit 31, and the desired driving voltage for each phase is converted.

<Operation of Position Detection Circuit> Next, the detailed operation of the position detection circuit 30 will be described. Position detection circuit 3
When 0, the position is detected by utilizing the inductance characteristic (see FIG. 16) that the inductance of the motor changes according to the rotor phase.

As shown in FIG. 2, the position detecting circuit 30 includes a current change amount detecting circuit 32, a rotational phase calculating circuit 34 and a differentiating circuit 35. The current change amount detection circuit 32 detects the current Iu, Iv, Iw of each phase of the motor winding detected by the current sensors 5u, 5v, 5w, and P from the three-phase PWM modulation circuit 31.
Using the change timing of the WM pulse signal, the current change rates δIu, δIv, and δIw per unit time and per unit voltage change (ie, normalized current change amounts) δI′u, δI′v, Find δI'w.

The rotation phase calculation circuit 34 calculates the current change rates δIu, δIv, and δIw in accordance with the below-described formula, and calculates the current change rates δI'u and δI which are the current change amounts per unit time and unit voltage change. The rotor position (rotational phase) θ ^{^} is obtained from'v, δI'w. The differentiation circuit 35 has a rotation phase θ ^{^.}
The velocity ω ^{^} is calculated by differentiating.

Here, a calculation formula for calculating the rotor position used in the rotation phase calculation circuit 34 will be described.

The current change rates δI'u, δI'v, δI'w obtained from the current change amount detection circuit 32 are closely related to the inductance of the motor winding of each phase. That is, the rate of change in current per unit time and unit voltage change is inversely proportional to the phase inductance.

FIG. 3 shows the relationship between the current waveform of the phase current and the voltage waveform of the instantaneous phase voltage. For convenience of description, only the U phase will be described. As shown in FIG. 3A, the instantaneous phase voltage δVu (the voltage obtained by subtracting the neutral point voltage, which is the average voltage of the phase from the phase terminal voltage, see FIG. 3D) becomes positive for the phase current. The period during which the instantaneous phase voltage is negative decreases. The amount of change in current δIu is proportional to the magnitude of this instantaneous phase voltage δVu, and is inversely proportional to the inductance of the motor winding. However, the inductance here (hereinafter referred to as "apparent inductance") is the combined inductance when the inductance of the corresponding phase itself and the other two phases connected in parallel are connected in series. . For example, the apparent inductance of the U phase is Lu ', and the inductance of each phase is Lu, Lv, Lw.
Then, it is calculated by the following formula. Lu ′ = Lu + (Lv // Lw) (1) On the other hand, the relationship between the phase inductance and the rotation phase θ ^{^} is as follows.

[Equation 4] This is replaced with Lu ', Lv', Lw 'described above, and further the inductances Lu', Lv ', Lw' and the current change amount δIu,
From the relationship with δIv, δIw, the current change rates δI′u, δI′v,
Rewriting to the equation of δI′w gives the following equation.

[Equation 5] Using the above equation, the rotational phase θ ^{^} can be obtained from the current change amounts δIu, δIv, δIw, that is, the current change rates δI′u, δI′v, δI′w.

FIG. 4 is a flow chart showing the procedure of position detection processing.
It is a chart. First, the current change amount detection circuit 32
The current values Iu, I of each phase are synchronized with the edge change of the phase current.
v and Iw are read (S251). Next, the previous edge change
From transition to current edge change δT and voltage in between
Normalize the current change amount δI using the change amplitude δV
The normalized current change amount, that is, unit time and unit
The current change rate per unit voltage change δI'u, δI'v, δI'w
Calculate (S252). The reason for normalization is that the edges are
Phase depends on the length of the period that does not change and the magnitude of the voltage amplitude.
Even if they are the same, the amount of change in the detected current is different.
It Next, the rotation phase calculation circuit 34
Current change rate per unit voltage change δI'u, δI'v, δI'w
From the above, the rotation phase θ^{^}Is calculated (S253).
Then, the differentiating circuit 35 rotates the rotation phase θ.^{^}To differentiate
More speed ω^{^}To calculate. Rotation phase
θ^{^}, Speed ω ^{^}Is required.

<PWM output pattern> The rotation phase can be obtained by obtaining the current change amount as described above, but in order to obtain the current change amount, it is necessary to detect the edge change of each phase current. . Therefore, the three-phase PWM modulation circuit 31 can reliably detect the edge change P
A processing procedure for generating the WM pulse will be described.

FIG. 5 is a flow chart showing a processing procedure for generating a PWM pulse that can reliably detect an edge change. The three-phase PWM modulation circuit 31 refers to the DC voltage Vdc from the three-phase voltage set values Vu *, Vv *, Vw * obtained by the coordinate conversion circuit 22 after the current compensation calculation by the compensation circuit 23, and then the PWM pulse The value of the basic duty (duty ratio) of is calculated (S201). Next, the calculated basic duty value is quantized with a duty ratio of 1/6 (that is, 0.16666 ...) (S202). At this time, the portion cut off by the quantization is stored in a predetermined storage area for later use.

Next, P is calculated using the quantized duty.
Referring to the WM pattern table (see FIG. 6), U,
The duty correction values of the V and W phases and the phase of the PWM pulse are obtained (S203). The PWM pattern table is stored in the three-phase PWM modulation circuit 31, the details of which will be described later. And for the correction value of the duty,
The values truncated when quantizing with a duty of 1/6 are added again to obtain the normal PWM timing of each phase (S
204). The obtained value is used as the pulse signal generating means PW.
Write to the M timer (corresponding to the output compare register buffer described later) (S205). Then, the edge generation time (phase) and the generation order of the edges of each phase are written in a predetermined storage means (S206), and the processing is ended. The information on the occurrence time and the occurrence order of the edges is referred to when detecting the current change of each phase.

The PWM pattern table will be described with reference to FIG. The table shown in FIG. 6 associates the basic duty of each phase of U, V, and W obtained after the compensation calculation and coordinate conversion with the output pattern of the PWM pulse for obtaining the output determined by the basic duty. Figure 6
Each PWM output pattern shown in is required according to the following rules. 1) Basically, the changing edges of each phase do not overlap (the timings of the changing edges are not the same). 2) For each phase, only one pulse is included in one carrier cycle, that is, one set of rising edge and falling edge is included in one carrier cycle. 3) Depending on the combination of basic duty, there are cases where it is not possible to generate a pattern in which edges do not overlap,
In that case, as an exception, some edges may overlap (see FIG. 6D).

In order to create the PWM output pattern in accordance with the above rules, a predetermined correction value ("Duty correction" in FIG. 6) is added / subtracted to / from one original basic duty value, if necessary. As described above, if the voltage difference between the phases does not change even if the predetermined value is added to each phase, a desired output defined by the basic duty can be obtained.

FIG. 6A shows P when the duty value of each phase of U, V and W is 1/6 (= 16.666 ...%), 1/6 and 1/6.
It is a WM output pattern. FIG. 6B shows a PWM output pattern when the duty value of each phase is 1/6, 1/6, 2/6. In the example of FIG. 6B, the duty values are added by +1/6 for each phase and corrected in order to prevent the edges from overlapping. Thus, even if the duty value is increased as a whole, the voltage difference between the phases does not change, so that a desired voltage can be obtained.

In FIG. 6C, the duty value of each phase is 1 /
It is a PWM output pattern in the case of 6, 1/6, and 3/6.

In FIG. 6 (d), the duty of each phase is 1/6,
It is a PWM output pattern in the case of 2/6 and 5/6. In this case, the duty is added by +1/6 for each phase for correction. In the example of FIG. 6D, as an exception,
There are places where edges overlap between the two phases. However, in this case, the timing when the edges of the two phases change at the same time is the current detection timing for the remaining phases whose edges do not change.

Depending on the combination of the input duty, there is a case where the 1/6 cycle quantization does not match. In that case, the correction amount is 1/12 duty. Further, FIG. 6 shows an example of the pattern, and in reality, it is necessary to prepare more output patterns according to the combination of the duty values of the respective phases. Further, in the present embodiment, when the quantization is performed, the duty is 1/6, but the duty may be 1/12, or the duty may be a multiple of 6.

The above-mentioned processing contents will be described with a specific example. For example, the DC voltage Vdc is 600V, and the voltage Vu of each phase is
*, Vv * and Vw * are 120V, 140V and 200 respectively
Consider the case of V. The basic duty of each phase is 1.2 /
6 (= 120/600), 1.4 / 6 (= 140/60)
0), 2/6 (= 200/600) (S20)
1). When the value of the basic duty of each phase is quantized by 1/6 duty and the fractions are rounded down, 1/6 and 1 /
It becomes 6 and 2/6 (S202). At this time, only 0.2 / 6 (= 20V) for the U phase and 0.
Only 4/6 (= 40V) is truncated by the quantization, and this truncated portion is stored. The PWM pattern table is referred to based on the duty value after quantization (S2
02). The duty value of each phase after quantization is 1/6, 1
/ 6 and 2/6, the duty of each phase is as shown in FIG.
The table of (b) is selected, and the generation timing (phase) and pulse width determined by the pattern of the table are determined (S203). Next, for the U phase and the V phase, 20 V and 40 V, which are rounded down at the time of quantization, are added to the respective duties (S204). The normal PWM timing thus obtained is written to the PWM timer (S205), and the edge occurrence time and the edge occurrence order are stored in a predetermined storage area (S206).
Here, in the case of FIG. 6B, the edge generation order is order information such as V phase rising, W phase rising, V phase falling, U phase rising, W phase falling, U phase falling. is there.

By using the duty value obtained as described above, the edge change of the phase current can be reliably detected. Therefore, a change in current indicating a change in inductance can be reliably detected, and the rotational phase θ can be reliably determined from the change in current according to the above-described equation (3).

<Initialization of Rotor Position> Referring to FIG. 16, even if one relationship between the inductances Lu, Lv, and Lw between the phases is determined, the relationship is 180 ° with respect to that relationship.
It can be seen that the two rotational phases that are offset correspond. Therefore, the rotational phase cannot be uniquely determined only from the inductance value. Therefore, in this embodiment,
By fixing the rotor at a predetermined position before starting the motor, it is possible to accurately grasp the rotor position after starting.

FIG. 7 is a flow chart showing a control procedure before the motor is started by the control circuit 7. In this control, the motor rotor is positioned at a predetermined position by energizing two predetermined phases (hereinafter, this processing is referred to as "motor initialization").

In FIG. 7, when the motor is started, first, energization is started only in two predetermined phases for initialization of the motor (S601). Then, it is checked whether or not a predetermined energization time has elapsed since the energization for initialization was started (S
602). If the predetermined time has not passed, the power supply to the predetermined two phases is continued (S603). At this time, the energization is performed so that the overcurrent does not flow while detecting the energization current. When the predetermined time has elapsed, the energization for initialization is stopped (S604), and the initialization is completed. The position of the rotor fixed by positioning is set to the initial value of the rotor position.

<Example of Hardware Configuration of Three-Phase PWM Modulation Circuit> FIG. 8 shows a three-phase PWM modulation circuit 3 for generating a PWM pulse signal.
It is the figure which showed an example of the hardware constitutions of 1.

The three-phase PWM modulation circuit 31 is the CPU 300
, Bus 308, UP counter 306, various registers 302, 303 ..., Comparator 301, AD
And a converter 351. In addition, the three-phase PWM modulation circuit 31
Includes circuit blocks 310, 320, 330 for generating PWM pulse signals for driving U-, V-, and W-phase switching elements in the three-phase PWM inverter circuit 4. Hereinafter, the circuit block 310 for the U phase will be described on behalf of the blocks 320 and 330 of each phase. The circuit block 310 for the U phase includes various registers 312, 31.
3, 316, 317 and the RS flip-flop 314
And comparators 311 and 315.

The UP counter 306 counts in synchronization with a predetermined high speed clock, and the count value is reset by a reset signal from the comparator 0 (301).

The CPU 300 has a register buffer (hereinafter referred to as “OCR-BF”) 30 via the bus 308.
Data for designating the timing of the edge of the PWM pulse can be written in 3, 313, and 317.

The OCR-BF-0 (303) has a CPU
By 300, the data that gives the reset timing of the UP counter 306, that is, the carrier period in the PWM control is written. The data written in the OCR-BF-0 (303) is transferred to the OCR-0 (302).

The initial value of the PWM pulse is written in the initial value buffer 304 by the CPU 300. The value of the initial value buffer 304 is reflected every time the UP counter 306 is reset.

OCR-BF-U (313), OCR-B
In the FX (317), the CPU 300 writes data that gives a timing for changing the edge of the U-phase switching element. That is, OCR-BF-U
The switching element of the upper arm of the U phase is set to O in (313).
The data that gives the timing to make N is OCR-BF-
Data that gives the timing for turning on the switching element of the lower arm of the U-phase is written in X (317).
OCR-BF-U (313), OCR-BF-X (31
The data written in 7) is the OCR-U (312),
Each is transferred to the OCR-X (316).

OCR-0 (302), OCR-U (31
2), to the OCR-X (316), the UP counter 306
The data is transferred from the corresponding buffer registers 303, 313, and 317 at the timing of resetting. That is, when the comparator-0 (301) detects that the count value of the UP counter 306 and the value of the OCR-0 (302) match each other, a reset signal is output, and each reset signal is output in synchronization with the reset signal. Register 3
Data is transferred to 02, 312, and 316. The UP signal 306 is reset, that is, initialized by the reset signal, and starts counting up until a match is detected again. That is, the UP counter 306 is initialized every carrier cycle. After that, OCR-BF-0 (30
Since the value written in 3) has been transferred to the OCR-0 (302), the coincidence detection pulse is repeatedly generated from the comparator-0 (301).

The coincidence detection pulse is OC which is another OCR.
It is also sent to the RU (312) and the OCR-X (316), and the OCR-BF- set by the CPU 300.
The values written in U (313) and PCR-BF-X (317) are respectively set to OCR-U (312) and OCR-X.
Transfer to (316). OCR-U (312), OCR
The output of -X (316) is connected to the comparator U (311) and the comparator X (315) and compared with the value of the UP counter 306. As a result of comparison, if the values of the UP counter 306 match, the comparators 311 and 31
A coincidence pulse is output from 5 and is input to the RS flip-flop 314 as a set signal and a reset signal, respectively. That is, the RS flip-flop 314 is O
It is set at the timing related to the value written in CR-BF-U (313), and OCR-BF-X (317)
It is reset at the timing related to the value written in. The output of the RS flip-flop 314 is the signal “U_O
N ”, and the inverted output signal is“ X_O
N ". These two signals are the three-phase PWM of FIG.
It is used to control the switching elements of the upper and lower arms of the U-phase of the inverter circuit 4.

Also, from the CPU 300 to the initial value buffer 3
Data is also written in 04. Initial value buffer 304
Is set as a preset value of the RS flip-flop 314 every time the comparator 0 (301) detects a match. That is, the RS flip-flop 314 is
The preset value is set each time the comparator-0 (301) detects a match, and each time a match between the comparator-U (311) and the comparator-X (315) is detected.
Set or reset.

The block 310 for the U phase has been described above, but the circuit blocks 320, 3 for the V and W phases are described.
The operation of 30 is similar.

The output signal of the RS flip-flop 314 is input to the AD converter 351 and the selection circuit 352 in addition to the control of the three-phase PWM inverter circuit 4. The AD converter 351 has the current information I selected by the selection circuit 352.
One of u, Iv, and Iw is AD-converted, and the value is stored. The value stored in the AD converter 351 is stored in the CPU
It can be read from 300. This allows three-phase P
The current value when the switching of the WM inverter circuit 4 is changed is automatically stored.

FIG. 9 is a timing diagram showing the operation of the circuit of FIG. The output of the UP counter 306 changes like a sawtooth wave. An interrupt request “IRQ_0” is issued to the CPU 300 at the timing of the sawtooth wave edge. C
The PU 300 uses various registers (O
(CR_0, OCR_U, OCR_X, etc.) is set.
The value set here becomes a valid value at the next IRQ_0.

If "1" is written in the initial value buffer 304, the U_ON output starts from "1" at the time of an interrupt request,
If "0" is written, the U_ON output starts from "0 (Low)". A counter value that gives a timing at which the U_ON output becomes “1 (High)” is written in OCR_BF_U (313), and OCR_BF_X (3
In 17), the counter value that gives the timing when the U_ON output becomes “0” is written. The set values written in these buffers become valid after the next IRQ_0.

<Another Configuration Example of Three-Phase PWM Modulation Circuit> FIG.
FIG. 9 is a diagram showing another configuration example of the three-phase PWM modulation circuit shown in FIG. 8. A difference from the circuit configuration of FIG. 8 is that a free-run counter 406 that counts in synchronization with a predetermined high-speed clock is used instead of the UP counter 306.
Since the free-run counter 406 is used, the coincidence output of the comparator 0 (301) is the free-run counter 4
It is not entered in 06. Other configurations and operations are the same as the circuit configuration of FIG.

FIG. 11 is a timing chart showing the operation of the circuit of FIG. The output value of the free-run counter 406 monotonically increases. Interrupt to CPU 300 "I
RQ_0 ”is generated by the comparator-0 (301) each time the counter value of the free-run counter 406 and the value of OCR-0 match. Therefore, if the data is written in the buffer register in consideration of the count-up value until the occurrence of the next interrupt every time, it can be used as in the case of the circuit of FIG. In this method, since the free-run counter is used, the count-up value information can be used for general purposes. That is, the free-run counter 406 can be used not only for main position detection but also for other control.

(Embodiment 2) Here, a position detection method capable of accurately detecting a position even when the power supply voltage drops for a short period will be described. In this method, for example, the capacity of the smoothing capacitor that smoothes the voltage after rectification is small,
This is an effective method in that position detection can be performed accurately even when the input voltage to the three-phase PWM inverter circuit drops to almost zero due to fluctuations in the power supply voltage.

FIG. 12 is a diagram for explaining the position detecting method in this embodiment. In the figure, time T
In the period from 1 to time T2, the input voltage (power supply voltage) abnormally drops and becomes almost zero.

FIG. 13 is a diagram showing a circuit configuration of the motor drive device of this embodiment. The overall configuration of the control circuit 7 is shown in FIG.
It is almost the same as that shown in. The difference is that an input voltage abnormality detection circuit 8 that detects an abnormality in the voltage of the single-phase power supply 1 is provided and the detection result is output to the control circuit 7, and P
Output terminal voltages Vu, Vv, Vw of the WM inverter circuit 4
Value (to be precise, voltage values Vu-n, Vv-n, Vw-n with reference to the negative side end potential of the rectifier circuit 3) is also input to the control circuit 7.

As shown in FIG. 12, when the input voltage drops at time T1, the input voltage abnormality detection circuit 8 transmits information indicating an abnormality to the control circuit 7. The input voltage abnormality detection circuit 8 determines that there is an abnormality when the input voltage Vac drops below a predetermined value. Based on this information, the control circuit 7 controls the PWM inverter circuit 4 to control the motor 2
When the power supply to the motor is stopped, the motor 2 operates as a generator and the generated voltage information is observed at the motor terminal. When each motor terminal voltage is observed with the negative side end potential of the rectifier circuit 3 as a reference, a mountain-shaped waveform having a flat section for a period of 120 degrees as shown in FIG. 12 is observed. Since the phase of the mountain-shaped waveform matches the phase of the phase induced voltage of the motor, the rotational position can be grasped by looking at the phase of the mountain-shaped waveform. When the waveform is approximated by the sin function, the U-phase induced voltage is highest at ± 60 deg with a peak at 90 deg.
Is a range of 30 deg to 150 deg.
Therefore, in the section where the abnormal input voltage is detected,
The rotor position can be predicted by detecting the terminal voltage of each phase and determining the magnitude relationship of the terminal voltages between the phases. The position detection of this embodiment utilizes this principle.

FIG. 14 is a flow chart showing the control for discriminating the rotational phase according to the magnitude relation of the motor terminal voltage, which utilizes this principle. This control is performed by the control circuit 7.

First, it is checked whether or not the input voltage is abnormal based on the output from the input voltage abnormality detection circuit 8 (S501).
If normal, normal control is performed (S510).

If it is abnormal, the power supply is turned off (S50
2), the magnitude relationship among the three-phase terminal voltages Vu, Vv, Vw is checked (S503). If "Vw>Vu>Vv", it is determined that the phase of the rotor is near 0 deg (S504). If "Vu>Vw>Vv", it is determined that the phase is in the vicinity of 60 deg (S505). If “Vu>Vv> Vw”, it is determined that the phase is near 120 deg (S506). If “Vv>Vu> Vw”, it is determined that the phase is 180 deg (S507). If "Vv>Vw>Vu", it is determined that the phase is near 240 deg (S508). If “Vw>Vv> Vu”, it is determined that the phase is near 300 deg (S509).

As described above, at the time of abnormality, the approximate value of the rotational phase can be detected by the magnitude order of the terminal voltage between each phase of the motor. Therefore, even if the power supply is restored, the approximate rotation phase can be captured, and thus the position sensorless drive can be immediately restored.

Since there is a monotonous correspondence between the rotational speed and the voltage required for driving, that is, the higher the rotational speed is, the higher the voltage required for driving is. It is preferable that the threshold value of the input voltage abnormality detection circuit 8 is variable with respect to the rotation speed.

[0069]

The first motor drive device or the first motor drive device of the present invention
According to the position detection method of (1), the rotation phase can be detected at any advance phase, without superimposing electric power other than for driving, on a carrier cycle basis without using a position sensor,
The effect that the advance value can be changed moment by moment is also obtained.

According to the second motor drive device or the second position detection method of the present invention, when a short-term abnormality occurs in the power supply, the power supply to the motor is stopped and the brake is generated. However, since the rough rotation phase can be captured, the rotation can be immediately and smoothly continued when the input voltage returns to the normal state.

[Brief description of drawings]

FIG. 1 is a block diagram showing an overall configuration of a motor drive device that uses a position detection device according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a detailed configuration of a control circuit according to the first embodiment.

FIG. 3 is a diagram showing a relationship between a phase current and a phase voltage of a motor.

FIG. 4 is a flowchart showing a procedure of a motor rotation phase detection process.

FIG. 5 is a flowchart showing a procedure for generating a motor drive PWM pattern.

FIG. 6 is a diagram showing an example of a PWM pattern table.

FIG. 7 is a flowchart showing a procedure of positioning processing when starting the motor.

FIG. 8 is a block diagram showing a configuration of a three-phase PWM modulation circuit according to a first example.

FIG. 9 is a diagram showing changes in the value of the UP counter and the value of the U_ON output of the three-phase PWM modulation circuit of the first example.

FIG. 10 is a block diagram showing a configuration of a three-phase PWM modulation circuit of a second example.

FIG. 11 is a diagram showing changes in the value of the free-run counter and the value of the U_ON output of the three-phase PWM modulation circuit of the second example.

FIG. 12 is a diagram for explaining a position detection method according to the second embodiment of the present invention.

FIG. 13 is a block diagram showing an overall configuration of a position detection device according to a second embodiment of the present invention.

FIG. 14 is a flowchart showing the operation of the control circuit in the position detecting device according to the second embodiment of the present invention.

FIG. 15 is a block diagram showing the overall configuration of a conventional motor drive device.

FIG. 16 is a diagram showing an inductance characteristic and an induced voltage characteristic of a motor which change according to a rotation phase.

FIG. 17 is a diagram showing a PWM output pattern that enables reliable detection of a phase current change edge in a conventional position detection method using inductance characteristics.

[Explanation of symbols]

1 ……………… Single-phase power supply 2 ……………… IPM motor 3 ……………… Rectifier circuit 4 …………… Three-phase PWM inverter circuit 5u, 5v, 5w …………… Current sensor 16… ………… Smoothing capacitors 7, 7b ………… Control circuit 8 ………… Input voltage abnormality detection circuit 20 ………… Torque control block 32 ………… Current change amount detection circuit 34 ………… Rotation phase calculation circuit 306 ... UP counter 301, 311, 315 ... Comparator 406 Free-run counter 351 ... AD converters 303, 304, 313, 317 ... Buffer register 314 ... ……… RS flip-flop

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5H560 BB04 BB07 DA14 DB20 DC12                       EB01 GG03 SS07 TT01 TT02                       TT04 TT07 TT10 TT11 UA02                       XA02 XA04 XA12 XA13                 5H576 BB06 BB09 CC05 DD05 EE01                       EE11 EE14 EE19 GG02 GG04                       HA02 HB02 JJ03 JJ12 JJ17                       JJ18 JJ19 JJ26 JJ29 LL14                       LL22 LL38 LL39 LL41

Claims (10)

[Claims] 1. A rectifying means for rectifying a single-phase AC voltage, and a DC voltage after rectification is converted into a pseudo three-phase AC voltage of arbitrary frequency and voltage by performing pulse width modulation (PWM) using a switching element. And an inverter means for
In a motor drive device that drives a motor with the pseudo three-phase AC voltage, means for obtaining an output voltage value to be output to each phase of the motor winding by a predetermined calculation, and a duty value and a PWM output to be output for the duty value A table in which patterns are associated with each other, the rising and falling edges of pulses do not overlap between phases,
Further, a table including a plurality of PWM output patterns determined so that only one pulse is included in one PWM carrier cycle, and the output voltage value is converted into a duty value in pulse width modulation, and the obtained duty value is Quantize with a resolution of a period that is an integral multiple of 1/6 of the PWM carrier cycle, select a PWM output pattern by referring to the table based on the quantized duty value, and output the PWM from the PWM output pattern. The duty value and phase of the pulse signal are determined, and the duty value of the normal PWM pulse signal for driving the switching element of the inverter means is added to the determined duty value by the amount cut off at the time of quantization. The pulse modulation means that determines the value and phase, and the current change of each phase of the motor winding is detected, and the unit time And a position detecting means for calculating the current rotation phase θ ^{^} using a predetermined calculation formula based on the current change rates δI'u, δI'v, δI'w which are the current change amounts per unit voltage change. A motor drive device characterized by the above. 2. The current change rate δI′u, δ according to the following equation:
The motor drive device according to claim 1, wherein the current rotation phase θ ^{^} is calculated based on I′v and δI′w. [Equation 1] 3. The motor is positioned by passing a constant current between two specific phases of the motor before the motor is started, and the phase at that time is set to the initial value of the rotor phase. Item 2. The motor drive device according to Item 1. 4. The pulse modulation means includes a counter for counting in synchronization with a predetermined clock and an initial value for storing a set value for determining a value of an output voltage for each switching element of the inverter means when a PWM pulse is generated. A value register, an initial value buffer register that stores data to be transferred to the initial value register at each PWM carrier cycle, and an output voltage value for each switching element from Low to Hig.
A first register that stores the counter value that rises to h, a first buffer register that stores the data to be transferred to the first register in each PWM carrier cycle, and a value of the output voltage for each switching element High to Lo
A second register that stores a counter value that falls to w, a second buffer register that stores data to be transferred to the second register in each PWM carrier cycle, and a value of an output voltage for each of the switching elements 2. The motor drive device according to claim 1, further comprising: a unit that stores a current value of the motor winding at a timing at which is changed. 5. The motor driving device according to claim 4, wherein the clock is initialized every PWM carrier cycle. 6. A means for detecting an input voltage of a single-phase alternating current, wherein a phase range is assigned in advance every 60 degrees in accordance with the magnitude relation between the voltages of the phases of the motor winding, and the input voltage is less than a predetermined voltage. When the voltage of each phase of the motor is detected, the terminal voltage of each phase of the motor is detected, and based on the magnitude relationship between the detected voltage of each phase of the motor winding, And a means for determining the phase of the rotor to a value of one value. 7. A single-phase AC voltage is rectified, and the rectified DC voltage is pulse width modulated (PWM) using a switching element.
A detection method for detecting the rotor position of the motor in the motor drive control for converting the pseudo three-phase AC voltage of arbitrary frequency and voltage to drive the motor using the pseudo three-phase AC voltage, An output voltage value to be output to each phase of the motor winding is obtained by a predetermined calculation, the output voltage value is converted into a duty value in pulse width modulation, and the obtained duty value is ⅙ of the PWM carrier cycle.
Quantizing with a resolution of a period of an integral multiple of, a table that associates the duty value and the PWM output pattern to be output for the duty value, the rising and falling edges of the pulses do not overlap between each phase,
And, referring to a table including a plurality of PWM output patterns determined so that only one pulse is included in one PWM carrier cycle, the PWM output pattern is selected based on the quantized duty value, and the PWM output is selected. Determine the duty value and phase of the PWM pulse signal that should be output from the pattern, and add the value that was truncated during quantization to the determined duty value to determine the duty value and phase of the regular PWM pulse signal. , The switching element is driven by using the determined regular PWM pulse signal, the current change of each phase of the motor winding is detected, and the current change rate δ which is the current change amount per unit time and unit voltage change is detected.
A method for detecting a motor rotor position, characterized in that a current rotational phase θ ^{^} is calculated using a predetermined calculation formula based on I'u, δI'v, δI'w. 8. The current change rate δI′u, δ is calculated by the following equation.
8. The motor rotor position detecting method according to claim 7, wherein the current rotation phase θ ^{^} is calculated based on I′v and δI′w. [Equation 2] 9. The motor is positioned by passing a constant current between two specific phases of the motor before the motor is started, and the phase at that time is set as an initial value of the rotor phase. Item 7. A method for detecting a motor rotor position according to Item 7. 10. A phase range is pre-allocated every 60 degrees according to the magnitude relationship between the voltages of the phases of the motor windings, the input voltage of a single-phase alternating current is detected, and the input voltage drops below a predetermined voltage. Is detected, the terminal voltage of each phase of the motor is detected, and based on the magnitude relationship between the detected voltage of each phase of the motor winding, A method for detecting a motor rotor position, characterized in that the phase of the rotor is determined as a value.