JP3353664B2 - Electrical angle detector - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a detection device for detecting an electrical angle in a synchronous motor, and more particularly, to a technology for detecting an electrical angle without using a dedicated sensor for detecting a rotational position of a rotor, that is, without using a sensor. .

[0002]

2. Description of the Related Art In order to obtain a desired rotational torque, a synchronous motor that rotates a rotor by flowing a multi-phase alternating current through a winding and an interaction between a magnetic field generated by the winding and a magnetic field generated by a permanent magnet requires a rotor. It is necessary to control the polyphase alternating current flowing through the windings according to the position, that is, the electrical angle. Conventionally, as a method of detecting the electrical angle of a rotor without a sensor, a device that applies a high-frequency voltage between windings and detects the rotor electrical angle from a voltage waveform between the windings has been proposed. This is based on the fact that in the case of a synchronous motor using a permanent magnet, a back electromotive force is generated between the windings due to the rotation of the rotor. On the other hand,
When the rotor is stopped or rotating at a low speed, even if the electromotive force is generated, the electromotive force is extremely small, so that the electrical angle cannot be accurately detected.

In view of such a problem, the applicant has proposed a device for accurately detecting the electrical angle of the rotor even when the rotor is stopped or rotating at a low speed (Japanese Patent Laid-Open No. 7-17).
7788). This applies a voltage between the windings of the motor,
The electrical angle is detected by detecting the behavior of the current flowing according to the applied voltage.

A specific example of the principle of detecting an electrical angle is as follows. Note that the electrical angle corresponds to θ in the equivalent circuit shown in FIG. FIG. 7 shows the relationship between the electrical angle and the current value when a relatively small voltage is applied to each phase in a three-phase motor composed of U, V, and W phases. The current of each phase is 1
Since the phases are shifted by 20 degrees, the magnitude relationship of the current flowing in each phase changes according to the electrical angle. Specifically, when the range of electrical angles from 0 to 180 degrees is divided into six sections of 30 degrees each, in sections 1 and 4 in FIG. 7, V phase, in sections 2 and 5, U phase, section 3 and section At 6, the W-phase current has an intermediate magnitude among the three phases (hereinafter, this current is referred to as “intermediate current”). As can be seen from FIG. 7, the current of each phase changes periodically, but the intermediate current can be approximated by a straight line. Accordingly, by comparing the magnitude relationship between the current values of the respective phases, it is determined which of the categories in FIG. 7 the electric angle belongs to, and then by using an approximate expression indicating the relationship between the current value of the intermediate current and the electric angle. , 0 to 180 degrees. On the other hand, electrical angle 1
Since a similar relationship is established in the range of 80 degrees to 360 degrees (sections 1 'to 6' in FIG. 7), the electrical angle can be similarly specified.

However, in this case, the electrical angle is 0 degree to 180 degrees.
In the section of degrees and the section of 180 to 360 degrees of electrical angle, it is not possible to specify which of the corresponding sections (for example, section 1 and section 1 ') the electrical angle is shifted by 180 degrees. Specifically, the state of the electrical angle of 30 degrees in section 1 and the state of the electrical angle of 210 degrees in section 1 'cannot be distinguished. Therefore, it is necessary to specify in which section the electrical angle belongs by separately determining the polarity of which of the N pole and the S pole of the magnet is closer to the U-phase coil.

As a method for specifying a section, for example, a method using magnetic saturation characteristics shown in FIG. 11 has been proposed. FIG. 11 shows the relationship between the magnetic flux H applied from the outside to the magnetic material forming the stator of the motor and the magnetic flux B inside the magnetic material. When the N pole of the permanent magnet attached to the rotor is located near the stator coil (FIG. 1)
At point a), a magnetic flux is applied to the magnetic body by a permanent magnet in advance. When the coil is energized in this state, the magnetic flux generated by the energization is added, so that the magnetic flux in the magnetic body enters a non-linearly changing region (point b in FIG. 11). On the other hand, when the S pole is located at a position close to the coil of the stator (point c in FIG. 11), the magnetic flux in the magnetic body remains in a region where the magnetic flux changes linearly even when the magnetic flux is applied by energizing the coil (FIG. 11). Point d).

[0007] Since such magnetic saturation characteristics affect the inductance of the coil and thus the detected current value, the peak value of the current after a voltage is applied for a certain period of time is measured, or the current is measured at a predetermined value. The positional relationship between the coil and the magnetic pole can be known by measuring the voltage application time required to reach the value. That is, whether the N pole is close to the coil (the electrical angle is in the range of 0 to 90 degrees or 270 to 360 degrees) or the S pole is close (90 to 270 degrees).
Degree section). As a result,
Of the two electrical angles detected 180 degrees apart earlier, the actual electrical angle is in the range of 0 to 180 degrees or 180 to 36 degrees.
It can be specified to which of the 0-degree sections it belongs, and the electrical angle can be uniquely detected in the range of 0 to 360 degrees.

[0008]

However, a section for determining whether or not magnetic saturation has occurred by measuring the peak value of the current or measuring the voltage application time required for the current to reach a predetermined value. In the determination method, it is very difficult to set a voltage pulse to be applied. That is, even when the positional relationship between the coil and the magnetic pole is in the state at the point a in FIG. 11, if the applied voltage is low, the magnetic flux density inside the magnetic body may not reach the saturated state. Therefore, even when the relationship between the coil and the magnetic pole is near the origin in FIG. 11, it is necessary to apply a voltage high enough to reach magnetic saturation. On the other hand, when the relationship between the coil and the magnetic pole is at the point c in FIG. 11, if a voltage that is too high is applied, the state passes the points d and a in FIG. 11 and magnetic saturation may occur. Further, it is necessary that the applied voltage value and the application time do not exceed the rating of the transistor inverter that controls the driving of the motor. The voltage pulse applied by the section determination method is determined experimentally. However, since the voltage pulse satisfying the above conditions is very limited, it is difficult to set an appropriate voltage pulse.

Further, the characteristics of the motor change due to the aging of the permanent magnet used for the rotor, the specification change at the time of designing the motor, and the like. Therefore, even if an appropriate pulse can be experimentally set as the applied voltage in the section determination method, it is necessary to set the voltage pulse again when the characteristic of the motor changes. Further, when a voltage fluctuation such as a voltage drop occurs in a battery that supplies power to the motor, there is a possibility that the above voltage pulse cannot be applied.

The present invention has been made to solve at least a part of the above problems, and has as its object to provide an apparatus capable of detecting an electrical angle without setting (tuning) fine voltage pulses.

[0011]

Means for Solving the Problems and Their Functions and Effects In order to solve at least a part of the above problems, the present invention employs the following means. The first electrical angle detection device of the present invention is:
An electric angle detecting device for a synchronous motor that rotates a rotor by an interaction between a magnetic field generated by a winding and a magnetic field generated by a permanent magnet, and applies a predetermined voltage to the winding. A voltage application unit, a current detection unit that detects a current behavior such as a rate of change of a current flowing according to the applied voltage, and an electrical angle of 0 to π based on the behavior of the current flowing according to the applied voltage. Alternatively, an electrical angle calculating means for calculating an electrical angle in a section of π to 2π, and an electrical angle of 0 to π or π to 2 based on the variation of the rate of change of the detected current.
and a section specifying means for specifying which section of π it belongs to.

According to the electric angle detecting device having such a configuration, the change rate of the current flowing according to the voltage applied to the winding of the synchronous motor can be detected, and based on the change in the current change rate, the electric angle can be detected. It specifies which section the angle belongs to from 0 to π or from π to 2π. The principle is as follows. In addition, since the electrical angle in the section of electrical angle 0 to π or π to 2π can be calculated,
The electrical angle can be detected in a range of up to 2π.

The section specifying means for specifying whether the electric angle belongs to a section from 0 to π or from π to 2π is based on the current value in that the specification is performed based on the change in the rate of change of the current. In principle, the specification is largely different from the prior art. The above-described magnetic saturation characteristics (FIG. 11) affect the current flowing through the winding. Normally, when a voltage is applied to the winding, a back electromotive force is generated in the winding in accordance with a change in the magnetic flux generated by the voltage, thereby preventing the current from flowing. This is a well-known effect of the winding inductance, and the current increases according to a curve determined by a function of the winding inductance. On the other hand, for example, when the magnetic flux density of the magnetic body approaches the saturation state as at point b in FIG. 11, the change in magnetic flux caused by the voltage applied to the winding is suppressed,
The back electromotive force generated in the winding is also reduced. As a result, a phenomenon occurs in which the current flowing through the winding rapidly increases. In other words, if a portion where the rate of change of the current flowing through the winding rapidly increases can be detected, it is detected that the magnetic flux density inside the magnetic body has reached a saturated state. In the present invention, using this principle, the saturation state of the magnetic flux density is detected based on the fluctuation of the current change rate, and it is specified which section of the electric angle belongs to 0 to π or π to 2π. .

According to such a principle, a voltage can be continuously applied until the magnetic flux density becomes saturated, so that the section can be specified without setting a voltage pulse in advance. Further, since it is possible to detect when the magnetic flux density becomes saturated after the start of the voltage application, in a specific state (for example, point c in FIG. 11), the voltage value and the voltage value are set so that the magnetic flux density does not become saturated. There is no need to limit the application time. Therefore, according to the electrical angle detection device of the present invention, the electrical angle can be detected without tuning the voltage pulse.

In the technical field of sensorless control of a motor, the relationship between the above-mentioned magnetic saturation characteristic and the current change rate has not been known so far. Was disclosed for the first time in the present application.

In the electrical angle detecting device, the voltage applying means is means for applying a predetermined voltage in one direction of the winding, and the section specifying means is configured to detect a detected current from the start of voltage application. Based on the magnitude relationship between the current rise time required for the change value of the rate of change to exceed a predetermined value and a predetermined threshold value related to a specific electric angle, the electric angle is 0 to π or π to 2π. It is desirable to use a means for specifying which section the image belongs to.

According to this configuration, it is possible to apply a voltage to the winding in one direction and detect a current rise time required until the detected fluctuation value of the current change rate exceeds a predetermined value. If the current rise time related to a specific electrical angle is set as a threshold, the detected electrical current rise time and, based on the magnitude relationship between the threshold and the electrical angle, the electrical angle can be any of 0 to π or π to 2π. It can be specified whether they belong. Further, since the current rise time can be detected only once, the section can be specified in a relatively short time.

The details are as follows. When the application direction of the voltage to the magnet coil winding belongs to the section where the electrical angle is 0 to π / 2 or 3 / 2π to 2π (hereinafter, referred to as N-polar section), the state belongs to the first quadrant of FIG. When the electrical angle belongs to a section of π / 2 to 3 / 2π (hereinafter referred to as an S-polarity section), the state belongs to the third quadrant in FIG. 11. Assuming that a voltage is applied to the winding in a direction in which the direction of the magnetic field due to the coil current coincides with the direction of the magnetic field due to the N pole at the electrical angle 0, as is apparent from FIG. Is shorter in the current rise time than in the case of belonging to the S polarity section. In other words, when the current rise time in the state where the electrical angle is π / 2 is set as the threshold, the current rise time when the electrical angle belongs to the N polarity section is smaller than the threshold, and the electrical angle is in the S polarity section. , The current rise time is longer than the threshold. Therefore,
By the magnitude relation between the current rise time and the predetermined threshold,
It is possible to specify to which section the electrical angle belongs from 0 to π or from π to 2π.

Although the above has been described in some detail for easy understanding, the present invention is not limited to these. For example, the direction in which the voltage is applied may be established in the opposite direction. In addition, the threshold value is an electrical angle of 0 to π.
Alternatively, it can be set according to the calculation section of the electrical angle calculation means for calculating the electrical angle in the section from π to 2π. For example, as described with reference to FIG. 7 in the related art, when the electrical angle calculation unit specifies the electrical angle within a unit of 30 degrees (π / 6), the threshold is set to the electrical angle. The current rise time in any state of π / 6 to π / 2 may be used.

On the other hand, in the electrical angle detecting device, the voltage applying means includes a forward voltage applying means for applying a predetermined voltage in one direction of the winding for a first predetermined period; And a reverse voltage application means for applying the same voltage as the predetermined voltage in the reverse direction for a second predetermined period, and the section identification means is detected from the start of voltage application by the forward voltage application means. The forward current rise time required for the variation value of the current change rate to exceed a predetermined value, and the time required for the detected variation value of the current change rate to exceed the predetermined value from the start of voltage application by the reverse voltage applying means. It is also preferable to use a means for specifying which of 0 to π or π to 2π the electrical angle belongs to, based on the magnitude relationship with the reverse current rise time.

With this configuration, it is possible to detect a forward current rise time by applying a voltage to the winding in one direction (forward direction), and to apply a voltage in the reverse direction to apply a reverse current rise time. Time can be detected. Since the magnitude of the forward current rise time and the magnitude of the reverse current rise time vary according to the electrical angle, the two are compared to determine whether the electrical angle belongs to a section from 0 to π or π to 2π. The electrical angle can be specified as 0 to 2
It can be detected in the range of π. Further, in this configuration, since the current rise time in the forward direction and the current rise time in the reverse direction are compared, there is an effect that the detection accuracy is stabilized as compared with a method of measuring the current rise time in one direction and comparing with a predetermined threshold value. . In other words, according to this configuration, section identification can be performed stably without being affected by changes in motor design specifications, fluctuations in battery voltage, aging of magnets forming the motor, and the like.

The section specifying method according to the above configuration will be specifically described as follows. When the electrical angle is in the state of the point a in FIG. 11, a voltage is applied to the winding in a direction (forward direction) where the direction of the magnetic field due to the coil current matches the direction of the magnetic field due to the N pole at the electrical angle 0. The forward current rise time is the time required to move to point b in FIG. On the other hand, when the voltage application direction is reversed,
The reverse current rise time is the time required to pass point c in FIG. 11 and further reach point e. Therefore, in this case, the forward current rise time is shorter than the reverse current rise time.

On the other hand, considering the case where the electrical angle is shifted by 180 degrees with respect to point a (state at point c in FIG. 11), the forward current rise time is the time required to pass point a and reach point b. Since the reverse current rise time is the time to reach the point e, the forward current rise time is longer than the reverse current rise time. From the above, the forward current rise time and the reverse current rise time are detected, and the magnitude relationship is compared.
Π or π-2π can be specified.

In this case, the first predetermined period of time in the forward voltage applying means is shorter than a period until the detected fluctuation value of the current change rate exceeds the predetermined value or a predetermined period. The second predetermined period in the reverse voltage applying means is either a period until the detected fluctuation value of the current change rate exceeds a predetermined value or a period exceeding the first predetermined period. The shorter period may be used.

In this way, since the voltage application period can be minimized, the section can be specified in a short period. First, a case where the fluctuation value of the current change rate exceeds a predetermined value during application of a forward voltage will be described. In this case, the forward current rise time can be measured when the variation value of the current change rate exceeds a predetermined value, and thereafter, no forward voltage is applied. Thereafter, when the fluctuation value of the current change rate detected during application of the reverse voltage exceeds a predetermined value, the reverse current rise time can be measured at that time, and thereafter, no reverse voltage is applied. In addition, when the fluctuation value of the current change rate does not exceed the predetermined value even after the period in which the forward voltage is applied, it can be determined that the reverse current rise time is longer than the forward current rise time at that time. Therefore, no reverse voltage is applied thereafter.

Next, a case will be described in which the fluctuation value of the current change rate does not exceed the predetermined value even if the application of the forward voltage exceeds a predetermined period. In this case, the application of the forward voltage is stopped when the predetermined period is exceeded. Thereafter, when the fluctuation value of the current change rate detected during application of the reverse voltage exceeds a predetermined value, it can be determined that the reverse current rise time is shorter than the forward current rise time at that time, Thereafter, no reverse voltage is applied.

As described above, by adopting the above configuration, it can be seen that the voltage application period can be minimized and the section can be specified in a short period. Here, the predetermined period is a period in which the fluctuation value of the current change rate can exceed the predetermined value in at least one of the case where the forward voltage is applied and the case where the reverse voltage is applied. For example, a period until the fluctuation value of the current change rate exceeds a predetermined value in a state where the electrical angle is π / 2,
The electric angle may be set in accordance with the calculation section of the electric angle calculation means for calculating the electric angle in the section from 0 to π or π to 2π.

[0028]

BEST MODE FOR CARRYING OUT THE INVENTION

(1) Configuration of Example Hereinafter, an embodiment of the present invention will be described using an example. FIG. 1 is a block diagram showing a schematic configuration of a motor control device 10 including an electric angle detection device as one embodiment of the present invention, and FIG. 2 is a three-phase synchronous motor 40 to be controlled.
FIG. 3 is a schematic diagram showing the general configuration of the three-phase synchronous motor 4.
FIG. 5 is an end view showing a relationship between a stator 30 of 0 and a rotor 50.

First, referring to FIG.
Will be described. This three-phase synchronous motor 40
Is composed of a stator 30, a rotor 50, and a case 60 for accommodating them. The rotor 50 has permanent magnets 51 to 52 affixed to its outer periphery, and a rotating shaft 55 provided at the center of the shaft is rotatably supported by bearings 61 and 62 provided in a case 60.

The rotor 50 is formed by laminating a plurality of plate-like rotors 57 formed by punching non-oriented electrical steel sheets. As shown in FIG. 3, the plate-like rotor 57 has four salient poles 71 to 74 at orthogonal positions. After laminating the plate-like rotors 57, the rotating shaft 55 is press-fitted, and the laminated plate-like rotors 57 are temporarily fixed. The plate-shaped rotor 57 made of this electromagnetic steel sheet has an insulating layer and an adhesive layer formed on its surface, and is heated to a predetermined temperature after lamination to be fixed by melting the adhesive layer.

After the rotor 50 is formed in this manner, permanent magnets 51 to 54 are attached to the outer peripheral surface of the rotor 50 at intermediate positions between the salient poles 71 to 74 in the axial direction. The permanent magnets 51 to 54 are magnetized in the radial direction of the rotor 50, and the polarity of adjacent magnets is different from each other. For example, the outer peripheral surface of the permanent magnet 51 has the N pole, and the outer peripheral surface of the adjacent permanent magnet 52 has the S pole.
It is a pole. The permanent magnets 51 and 52 are
0 is assembled to the stator 30, the plate-like rotor 57
And a magnetic path Md penetrating the plate-shaped stator 20 (FIG. 3).
See broken line).

The plate-like stator 20 constituting the stator 30 is
Like the plate-like rotor 57, it is formed by punching a thin sheet of non-oriented electrical steel sheet, and as shown in FIG.
The teeth 22 are provided. A coil 32 for generating a rotating magnetic field in the stator 30 is wound around a slot 24 formed between the teeth 22. In addition, a bolt hole through which a fixing bolt 34 is passed is provided at an outer edge portion of the plate-shaped stator 20, but is not shown in FIG. 3.

The stator 30 is temporarily fixed by heating and melting the adhesive layer in a state where the plate-like plate-like stators 20 are stacked and pressed against each other. In this state, the coil 32 is wound around the teeth 22 to complete the stator 30, which is assembled to the case 60, and the fixing bolt 34 is inserted into the bolt hole.
And tighten it to secure the whole. Further, the three-phase synchronous motor 40 is completed by rotatably assembling the rotor 50 with the bearings 61 and 62 of the case 60.

When an exciting current is applied to the coil 32 of the stator 30 so as to generate a rotating magnetic field, a magnetic path Mq penetrating the adjacent salient poles and the plate-like rotor 57 and the plate-like stator 20 as shown in FIG. Is formed. Note that an axis through which the magnetic flux formed by the permanent magnet 51 penetrates the rotor 50 in the radial direction is called a d-axis, and an axis through which the magnetic flux formed by the coil 32 of the stator 30 passes through the rotor 50 in the radial direction. Call it the q-axis. The angle that the d-axis electrically forms with the q-axis is called an electrical angle. FIG. 4 shows an equivalent circuit of the three-phase synchronous motor of this embodiment. As shown in FIG. 4, the electrical angle corresponds to the angle between the U-phase coil and the magnetic pole in the equivalent circuit.

Next, the configuration of the motor control device 10 will be described with reference to FIG. The motor control device 10 controls the three-phase (U, V, W phase) motor current of the three-phase synchronous motor 40 in response to an external torque command.
100, current sensors 102, 10 for detecting the U-phase current Au, the V-phase current Av, and the W-phase current Aw of the three-phase synchronous motor 40
3, 104, filters 106, 107, 108 for removing high frequency noise of the detected current, and two analog-to-digital converters (ADCs) 112, 113, 114 for converting the detected current value into digital data. I have.

As shown, a microprocessor (C) for performing arithmetic and logical operations is provided in the control ECU 100.
PU) 120, a ROM 122 preliminarily storing the processing and necessary data performed by the CPU 120, a RAM 124 for temporarily reading and writing data required for the processing, a clock 126 for clocking, and the like. Have been. The input port 116 and the output port 118 are also connected to this bus, and the CPU 1201
Through these ports 116 and 118, currents Au, Av and A flowing in the U, V and W phases of the three-phase synchronous motor 40, respectively.
w can be read.

Further, the motor current control circuit 100 includes the motor phase currents Au, determined based on the torque command.
A voltage application unit 130 that applies a voltage between the coils of the motor so as to obtain Av and Aw is provided at the output unit. The control output from the CPU 120 is output to the voltage application unit 130, and the voltage applied to each coil of the three-phase synchronous motor 40 can be controlled from outside.

It should be noted that, among the configurations of the motor control device 10 described above, the current sensors 102 to 104 and the filters 106 to
8. The ADCs 112 to 114 and the control ECU 100 correspond to an electric angle detecting device.

(2) Electric Angle Detecting Process An electric angle detecting routine by the electric angle detecting device 10 of this embodiment will be described with reference to FIG. The electric angle detection routine is a routine periodically executed by the CPU 120 together with various routines for controlling the operation of the motor 40. In the following description, electrical angles of 0 to π
Alternatively, the electrical angles calculated in the range of π to 2π are called electrical angles φ1 and φ2 (both are collectively called φ), and the electrical angles specified in the range of electrical angles 0 to 2π are electrical angles θ. Call and distinguish. When the electric angle detection routine is started, the CPU 120 performs a pulse pattern determination process (Step S100). Here, the contents of the pulse pattern determination processing will be described with reference to FIG.

In the pulse pattern determination process, the CPU 120
First detects the U-phase current Au (step S10).
5). Specifically, CPU 120 includes a voltage application unit 130
And a relatively low voltage that does not cause magnetic saturation is applied between the U-VW phases. At a point in time when a predetermined period has elapsed from the voltage application, the output of the current sensor 102 is passed through the filter 106 and the ADC 112. Read as U-phase current. Next, the CPU 120 performs the same processing to output the voltage V through the current sensor 103, the filter 107, and the ADC 113.
A phase current is detected (step S110). Further, by the same processing, the current sensor 104, the filter 108, the AD
The W-phase current is detected via C114 (step S11)
5). U, V, W phase currents Au, A
The electrical angle existence category (n) is determined based on v and Aw (step S120), and the electrical angle φ1 is calculated (step S125). This determination will be described with reference to FIG.

The current Au detected in the above steps,
FIG. 7 shows a state of change of Av and Aw. The current between the phases is shifted from the electrical angle of 0 degree because the phases are shifted by 120 degrees.
Focusing on 180 degrees, the magnitude relation of the current of each phase is 30
It can be seen that it changes by degrees and the following six combinations are obtained in total. Category 1 (0 to 30 degrees) → Au>Av> Aw Category 2 (30 to 60 degrees) → Av>Au> Aw Category 3 (60 to 90 degrees) → Av>Aw> Au Category 4 (90 to 120 degrees) → Aw>Av> Au classification 5 (120-150 degrees) → Aw>Au> Av classification 6 (150-180 degrees) → Au>Aw> Av

Of the phase currents in each section, those having an intermediate magnitude (for example, Av in section 1, Au in section 2)
.) And the electrical angle φ1 can be approximated by regarding this section as a straight line. That is, the current average value of each phase is Aav, and the current that is linearly approximated by the nth section is A
If n and the inclination of the straight line are m, the electrical angle φ1 is given by the following equation (1).
Is represented by φ1 = (n−1) × 30 + 15 + sgn (Aav−A
n) × m (1) Here, sgn is a function that takes a value of 1 for categories 1, 3, and 5, and a value of −1 for categories 2, 4, and 6. In addition, An = Av in sections 1 and 4, An = Au in sections 2 and 5, and An = Aw in sections 3 and 6. These relationships also hold for electrical angles of 180 to 360 degrees.

As a specific example, Au = 130A, Av = 95
A, Aw = 112A and m = 1.5 will be described. At this time, the magnitude relation of the current of each phase is Au>Aw>
Since it is Av, it is understood that the electrical angle φ1 belongs to the section 6. Therefore, in the above equation (1), n = 6, s
By substituting gn = -1 and Aav = 12.3A, the electrical angle φ1 = 165 degrees is obtained. Paying attention to the electrical angle of 180 to 360 degrees, the same relationship is established in the section 6 ′, so that the electrical angle φ2 = 345 degrees is obtained. That is, at this stage, the electrical angle is not uniquely determined in the range of 0 to 2π.

In the specific example described above, the current values Au, A actually measured as the average current Aav used for calculating the electrical angle are described.
Although the average value of v and Aw is used, when the average value Aav is obtained from the control, it may be calculated using that value. Alternatively, the current value may be measured, for example, only for Au and Av, and Aw may be calculated backward from Aav.

Further, the determination of the electric angle existence category (n) (step S120) and the calculation of the electric angle (step S1)
Regarding 25), as shown in FIG. 8, the following method is also possible. In the case of a cosine wave signal, it is known that an approximate expression of θ ≒ (tan2θ) / 2 holds when the phase is near the angle 0. Maximum current Au, Av, Aw for each phase
Since the phase is shifted by 120 degrees, the following equation (2) is obtained by expanding this approximation equation for each phase current. θ ≒ (tan2θ) / 2 Right-side numerator = sin2θ = {3 (IB-IC) Right-side denominator = 2cos2θ = 2 {2IA- (IB + IC)} = 6IA (2) where IA is a pole near phase 0 degree. It is the deviation between the interphase current taking a value and the average value Aav, and IB and IC are the deviations between the other interphase currents and the average value Aav. In addition,
When one interphase current has an extreme value, the sign (±) of the deviation from the average value of the other two interphase currents is the same. The relationship between the signs of the deviations △ Au, △ Av, △ Aw from the average value Aav of the inter-phase currents Au, Av, Aw and the divisions is shown below.

Category ΔAu ΔAv ΔAw Category 1a (-15 to 15 degrees) Positive (extreme value) Negative Negative Category 2a (15 to 45 degrees) Positive Positive Negative (extreme value) Category 3a (45 to 75 degrees) Negative Positive (extreme value) Negative Category 4a (75 to 105 degrees) Negative (extreme value) Positive Positive Category 5a (105 to 135 degrees) Negative Negative Positive (extreme value) Category 6a (135 to 165 degrees) Positive Negative (extreme value) Correct Therefore, as shown in FIG. 8, the section 1a in this embodiment
Through 6a are each 30 degrees starting from -15 degrees. FIG. 8 shows a portion used for approximation calculation within the section.
Are shown in bold lines.

The electrical angle φ can be obtained by performing an operation for each of the above-described six sections by using an equation obtained by replacing △ Au, △ Av, and △ Aw in the above equation (2). Equation (5) is
Since the phase is an approximate expression near 0 degrees, 30
A process of adding each (× (n−1)) times is also required. The approximate expression for each section is as follows: A = △ Av− △ Aw, B = △ Au−
Defined as ΔAv, C = ΔAw−ΔAu, and the approximate approximation formula 1a {3A / 6} Au 2a 30 + √3B / 6 △ Aw 3a 60 + √3C / 6 △ Av 4a 90 + √3A / 6 △ Au 5a 120+ {3B / 6} Aw 6a 180+ {3C / 6} Av. Similarly, the electric angle φ2 can be calculated in the electric angle range of 180 degrees to 360 degrees.

Returning to FIG. 6, the contents of the pulse pattern determination processing will be described. The electric angle existence class (n) is obtained by the above-described processing.
(Step S120) and the calculation of the electrical angle φ (step S125), the CPU 120 determines a pulse pattern for applying a voltage to the winding based on the electrical angle existence category (n) (step S130). Specifically, the electrical angle existence class (n) is defined as a phase between which a voltage is applied.
Select the phase giving the maximum current value in. In the case where the electric angle φ is calculated by the method shown in FIG. 7, when the electric angle belongs to the section 6 or the section 6 ′ as in the specific example described above, the U-phase among the three phases is used. Since the current value is maximum, the CPU 120 selects between the U-VW phases as the phase to which the voltage is to be applied. This means that, in other words, in the equivalent circuit diagram shown in FIG. 4, the phase in which the angle between the coil winding and the magnetic pole is minimized is selected. Thus, by selecting the phase to which the voltage is applied, the electrical angle can be efficiently and stably detected.

After executing the pulse pattern determination processing described above, the CPU 120 returns to the electrical angle detection routine of FIG. 5, and executes the forward current rise time (T1) measurement processing in the next step (step S105).
As described in the above specific example, in the pulse pattern determination processing, the electrical angle is not uniquely determined in the range of 0 to 2π. In order to uniquely determine the electrical angle, it is necessary to determine which section of the electrical angle φ belongs to 0 to π or π to 2π (hereinafter, polarity determination).
Forward current rise time measurement processing (step S10
5) is one process for polarity determination. The contents of this processing will be described with reference to FIG.

When the forward current rise time (T1) measurement processing routine is started, the CPU 120 detects an initial current A0 (step S205). Here, the current value A0 is the current value of the phase to which the voltage is to be applied, which is selected in the pulse pattern determination process (step S100). That is, when a voltage is selected to be applied between the U-VW phases, the current value A0 is a U-phase current value.
At the time when the initial current A0 is detected, no voltage is applied to the coil winding between any phases.
The CPU 120 substitutes the value α for the current change rate △ A0 to detect the initial current A0 and initialize the current change rate △ A0 (step S205). The value α may be a value sufficiently larger than a value ε described later.

Next, the CPU 120 starts the application of the forward voltage and substitutes a value 0 into a timer T1 for measuring the elapsed time from the start of the voltage application to reset (S210). The forward voltage refers to a state in which, when a voltage is applied between the U-VW phases, the U phase is set to a positive voltage, and the V and W phases are set to the ground voltage. The CPU 120 outputs a control signal to the voltage application unit 130 so that the voltage is applied as described above. FIG. 10 shows an example of a voltage pulse when a forward voltage is applied between the U-VW phases. The portion indicated by pulse pattern # 1 in FIG. 10 is the forward voltage. In this part, the U-phase voltage Vu has a positive value, and the V and W-phase voltages Vv and Vw are in a negative state.

In the next step, the CPU 120 sets the timer T
It is determined whether 1 is smaller than a predetermined limit value TL (step S215). If the timer T1 is equal to or greater than the limit value TL, that is, if the forward voltage application time has reached TL, the CPU 120 temporarily stops the application of the forward voltage (step S245), and performs an electrical angle detection routine (FIG. 5). Return to). At this time, the timer T1 has a value equal to or greater than the limit value TL.

If the timer T1 is smaller than the limit value TL, the CPU 120 waits until the time Ts elapses and increases the value of the timer T1 by Ts (step S2).
20). The time Ts is a sampling time for detecting the current, and is set experimentally to a value at which the change rate of the current flowing through the coil winding can be sufficiently detected.
Waiting for the elapse of the time Ts (step S220)
While the PU 120 executes another routine for controlling the operation of the motor, this routine is executed by executing an interrupt process every time Ts and executing this routine.

After the elapse of the sampling time Ts, C
The PU 120 detects the current value A1 (Step S22)
5), Calculate the current change rate ΔA1 (Step S23)
5). The current value A1 to be detected is a current value in the same phase as the initial current A0. The current change rate ΔA1 is the difference between the current values A1 and A0 (ΔA1 = A1−A0). Since the sampling time Ts is constant during the execution of this routine, it is not necessary to divide the above difference by the sampling time Ts to obtain the current change rate ΔA1. If the sampling time Ts is likely to change during execution of this routine, the current change rate ΔA1 needs to be a value obtained by further dividing the difference by the sampling time Ts.

After calculating the current change rate in this way, the CPU
120 is the current change rate ΔA1 and the initial current change rate ΔA0
(△ A1- △ A0), that is, whether or not the variation value of the current change rate is larger than a predetermined deviation ε (step S240). When the variation value of the current change rate is equal to or smaller than the predetermined deviation ε, the current value A0 and the current change rate △ A0 are replaced with A1 and △ A1, respectively (step S230),
The process returns to step S215 and subsequent steps. That is, after the elapse of the sampling time Ts, the calculation of the current change rate and the like are performed. On the other hand, the fluctuation value of the current change rate is equal to the predetermined deviation ε.
If it is larger than this, the application of the forward voltage is stopped (step S245), and the present routine is ended once.

This will be described with reference to FIG. FIG.
The state of the U-phase current in the pulse pattern # 1 during 0 is Au
様 子 Au shows the state of the current change rate of the U phase. FIG.
A broken line in 0 indicates a sampling timing of current detection. As shown by Au in FIG. 10, the U-phase current rises in a state relatively close to a straight line immediately after the start of voltage application (t11 in FIG. 10), but rises sharply after time t15. This is because the magnetic saturation (the state of the point b in FIG. 11) of the teeth 22 around which the coil winding is wound occurs at the time t15.

Current change rate ΔA in pulse pattern # 1
As for u, the result calculated by the CPU 120 is plotted. During time t11 to t12, Au in FIG.
, The current has changed, but since the CPU 120 has not detected this current value, the current change rate △ Au has a value of 0.
Remains. At time t12, the current change rate 電流
Au is calculated. The values here are the time t11 and the time t
12 is a value calculated using the current value in FIG. In addition, the variation value δ1 of the current change rate at this time is a predetermined deviation ε.
It is a fluctuation value smaller than. The current change rate △ Au takes a constant value until time t16 because the current Au increases linearly. Although the current sharply increases between times t15 and t16, for the same reason as described above, no change appears in the current change rate △ Au, and the current rapidly increases at time t16. The variation value δ of the current change rate here
2 is a value larger than the predetermined deviation ε. Therefore, CP
U120 stops the forward voltage application at time t16.

From the above, the forward current rise time (T
1) In the measurement processing routine, the time from the start of voltage application in the forward direction to the time when magnetic saturation occurs in the teeth 22 around which the coil winding is wound and the current flowing through the coil sharply increases (rises) is measured. It is understood that it is done. on the other hand,
Even if the elapsed time after application of the forward voltage reaches the predetermined limit value TL, if the current change rate does not fluctuate beyond the predetermined deviation ε, the forward current rise time (T1)
Stores a time that slightly exceeds the predetermined limit value TL.

The values of the predetermined deviation ε and the like used in the above description are experimentally obtained so that the rise of the current Au can be appropriately detected. For example, the deviation ε is a value in a range that is smaller than the fluctuation value of the current change rate at the time of the rise of the current (δ2 in FIG. 11) and larger than the fluctuation values of the other current change rates (δ1 in FIG. 11).

The limit value TL of the voltage application time is a value slightly exceeding the maximum time required for magnetic saturation to occur. This value needs to be set in consideration of the applied voltage value and the range of the electric angle of the motor. In the present embodiment, the pulse pattern determination processing described above (step S100 in FIG. 5)
, It is determined which of the sections (sections 1 to 6 in FIG. 7) the electrical angle belongs to by 30 degrees, and the polarity determination is that the electrical angle φ is in the range of positive and negative 30 degrees. It may be performed in a state. Therefore, the voltage application time limit value T
L is when the electrical angle is slightly over 30 degrees,
The time required until magnetic saturation occurs in the corresponding magnetic pole may be set.

It is to be noted that the limit value TL is provided because, for example, when the S pole is opposed to the U phase, it takes a long time until the current rises by applying a voltage in the forward direction. This is to avoid a time delay.
Therefore, if it is not necessary to seek such an effect,
The determination in step S215 may be deleted, and the application of the voltage may be continued until the current rises.

Next, the CPU 120 returns to the electrical angle detection processing and executes a reverse current rise time (T2) measurement processing (step S300 in FIG. 5). The reverse current rise time (T2) measurement process is also one process for polarity determination. The contents of this processing will be described with reference to FIG. Although not shown in the flowchart of FIG. 5, the reverse current rise time measurement processing (step S300)
Forward current rise time measurement processing (step S20)
0) is completed, and is executed after the current flowing through the coil winding has sufficiently attenuated.

When the reverse current rise time (T2) measurement processing routine is started, the CPU 120 determines the same phase (U phase in the example of FIG. 10) as the phase measured in the forward current rise time (T1) measurement processing. The initial current a0 is detected, and the value α is substituted into the current change rate △ a0 to perform initialization (step S305). The value α may be a value sufficiently larger than the value ε described later, and the forward current rise time (T
1) It does not need to be the same value as the value α in the measurement processing (step S205 in FIG. 9).

Next, the CPU 120 starts the application of the reverse voltage, substitutes a value of 0 into a timer T2 for measuring the elapsed time from the start of the voltage application, and resets it (S31).
0). The reverse voltage is a voltage in the reverse direction with respect to the forward voltage described above. When a voltage is applied between the U and VW phases, the U phase is set to the ground voltage and the V and W phases are set to the positive voltage. Say. The portion indicated by pulse pattern # 2 in FIG. 10 is the reverse voltage. In this part,
The U-phase voltage Vu is in the ground state, and the V and W-phase voltages Vv and Vw have positive values. Note that the absolute value of the applied voltage is the same as the voltage value applied in the forward current rise time measurement process.

In the next step, the CPU 120 sets the timer T
It is determined whether or not 2 is shorter than the forward current rise time T1 (step S315). Timer T2 is time T
If it is 1 or more, that is, if the application time of the reverse voltage exceeds the forward current rise time T1, the CPU 12
0 temporarily stops application of the reverse voltage (step S34).
5) Return to the electrical angle detection routine (FIG. 5). At this time, the timer T2 has a value equal to or longer than the forward current rise time T1.

In step S315, unlike the forward current rise time measurement process (FIG. 9), the limit value TL is set.
And did not compare. As described later, the measurement of the forward and reverse current rise times T1 and T2 is performed to compare the magnitudes of the two.
This is because the magnitude relationship between the two becomes clear when the reverse current rise time T2 exceeds the forward current rise time T1. That is, in step S315, T2 and T
Performing the comparison of 1 shortens the time required to execute this routine. However, if such an effect is not considered, the limit value TL and the timer T2 may be compared in step S315 as in the forward current rise time measurement processing (FIG. 9).

If the timer T2 is smaller than the time T1, the CPU 120 waits for the time Ts to elapse and increases the value of the timer T2 by Ts (step S3).
20). Further, the current value a1 is detected (step S32).
5), calculate the current change rate △ a1 (step S33)
5). After calculating the current change rate in this manner, the CPU 120
Determines whether the variation value (値 a1- △ a0) of the current change rate is larger than a predetermined deviation ε (step S34).
0), when the fluctuation value is equal to or smaller than the predetermined deviation ε, the current value a0 and the current change rate △ a0 are set to a1 and △ a1 respectively.
(Step S330), and again at step S31
The process returns to step 5 and subsequent steps. That is, the sampling time T
After elapse of s, calculation of the current change rate and the like are performed. on the other hand,
If the fluctuation value is larger than the predetermined deviation ε, the application of the reverse voltage is stopped (step S345), and the present routine is ended once. The processing in these steps is performed in steps S220 to S220 of the forward current rise time measurement processing.
This is the same as the process in H.245. Note that the predetermined deviation ε
Is the same value as the deviation ε in the forward current rise time measurement processing.

This will be described with reference to FIG. FIG.
In the 0 pulse pattern # 2, as shown by Au, the U-phase current rises in a relatively linear state immediately after the start of voltage application (t21 in FIG. 10), but after time t23, the magnetic saturation of the teeth 22 (see FIG. State of point e in 11)
, The current rises rapidly. Since the application direction of the voltage is opposite, the current value Au appears as a negative current. Hereinafter, the current value Au and its change rate △ Au are all treated as absolute values.

Current change rate ΔA in pulse pattern # 2
In the state of u, the current change rate △ Au is calculated at time t22, and the fluctuation value δ3 is a fluctuation value smaller than the predetermined deviation ε. Since the current Au increases linearly, the current change rate 時刻 Au takes a constant value until time t24, and rapidly increases at time t24. Here, the fluctuation value δ4 of the current change rate is a value larger than the predetermined deviation ε. Therefore, the CPU 120 stops applying the voltage in the reverse direction at the time t24.

As described above, the reverse current rise time (T
2) In the measurement processing routine, the time from the start of voltage application in the reverse direction to the time when magnetic saturation occurs in the teeth 22 around which the coil winding is wound and the current flowing through the coil rapidly increases (rises) is measured. It is understood that it is done. on the other hand,
Even if the elapsed time after the application of the reverse voltage reaches the forward current rise time T1, if the current change rate does not fluctuate beyond a predetermined deviation ε, the reverse current rise time (T2) Means that a time slightly exceeding the time T1 is stored.

Returning to FIG. 5, the electrical angle detection routine will be described. After the reverse current rise time measurement processing (Step S300) is completed, the CPU 120 compares the magnitude of the forward current rise time T1 with the magnitude of the reverse current rise time T2 (Step S400). The polarity is determined based on the magnitude relationship, and one of the electrical angles φ1 and φ2 calculated at the electrical angles 0 to π and π to 2π in the previous pulse pattern determination process (step S100) is selected as a correct value. The electrical angle θ is uniquely determined within the range of 0 to 2π.

If the time T1 is longer than the time T2, the electric angle calculated in the electric angle range of 0 to π according to the section to which the electric angle φ determined in the pulse pattern determination processing (step S100) belongs. Angle φ1 or electrical angle π
The electrical angle φ2 calculated in the range of 2π to 2π is selected as follows, and the electrical angle θ is determined in the range of 0 to 2π. When it is determined that the electrical angle φ belongs to the sections 1 to 3 or the sections 1 ′ to 3 ′, the electrical angle φ2 calculated in the range of the electrical angle π to 2π is selected, and the electrical angle θ = φ2 (Step S4
05). When it is determined that the electrical angle φ belongs to the sections 4 to 6 or the sections 4 ′ to 6 ′, the electrical angle φ1 calculated in the range of the electrical angles 0 to π is selected, and the electrical angle θ = φ1 (Step S405).

On the other hand, when the time T1 is equal to or shorter than the time T2, the electrical angle θ is obtained from the reverse relationship. When it is determined that the electrical angle φ belongs to the sections 1 to 3 or the sections 1 ′ to 3 ′, the electrical angle φ1 calculated in the range of the electrical angles 0 to π is selected and the electrical angle θ = φ1 (Step S410). Electrical angle φ is 4-6 or 4'-6 '
If the electrical angle is determined to belong to
The electrical angle φ2 calculated in the range of π is selected, and the electrical angle θ = φ
2 (step S410).

The reason why the electrical angle θ is determined by such a determination will be described with reference to FIG. FIG. 13 shows FIG.
3 shows the relationship between the U phase and the magnetic pole position, that is, the electrical angle in the equivalent circuit of the three-phase synchronous motor 100 shown in FIG. N
When the pole is in the state facing the U phase, the electrical angle is 0 degree, and therefore, the sections 1 to 6 and the sections 1 ′ to 1 shown in FIG.
6 'corresponds to the case where the magnetic poles are in the range shown in FIG.

Next, the relationship between the magnetic pole position and the forward current rise time T1 and the reverse current rise time T2 in FIG. 13 will be described. Time T1 is the time from the start of voltage application until magnetic saturation occurs when a forward voltage is applied to the U phase, and time T2 is the time from when magnetic saturation occurs by applying a reverse voltage to the U phase. The time required for Therefore, when the time T1 is shorter than the time T2, the N pole is close to the U-phase coil (the hatched area in FIG. 13), and when the time T1 is longer than the time T2, the S pole is U State close to phase coil (Fig. 13
In the area without hatching). In other words, when the time T1 is shorter than the time T2, the electrical angle θ
Belongs to sections 1-3 or sections 4′-6 ′, and conversely, the electrical angle θ is section 1′-3 ′ or sections 4-6.
It belongs to. From the above, it can be seen that the electrical angle θ is determined in steps S405 and S410 according to the relationship between the magnitude relationship between the times T1 and T2 and the section to which the electrical angle φ belongs.

However, in the present embodiment, since the phase to which the voltage is applied is selected in the pulse pattern determination processing (step S100) so that the coil winding and the magnetic pole exist in the closest section, the above determination is further simplified. It becomes something. Specifically, the relationship between the section to which the electrical angle φ belongs and the phase to which the voltage is applied is as follows. Electric angle exists in sections 1, 6, 1 ', 6' → Voltage applied to U phase Electric angle exists in sections 4, 5, 4 ', 5' → Voltage applied to V phase Electric angle applied to sections 2, 3, 2 ′, 3 ′ → Apply voltage to W phase Therefore, for example, when voltage is applied to the U phase, in steps S405 and S410, sections 1 and 1 ′ or sections 6 and 6 ′ By judging to which of these categories the electric angle θ can be determined. The same applies when a voltage is applied to the V phase and the W phase.

According to the embodiment described above, the electrical angle can be stably detected without requiring fine adjustment of the voltage pulse applied to the coil winding. According to the present embodiment, the polarity is determined based on the change in the rate of change in current flowing when a voltage is applied. Even if the voltage value is low, magnetic saturation occurs due to continuous application of the voltage. is there. In this embodiment as well, it is necessary to experimentally set the voltage value and the like to be applied.
In this embodiment, the upper limit value of the voltage is determined so that a current exceeding the rating of 0 does not flow, or the voltage is set such that the current change rate can be appropriately detected from the relationship with the sampling time of the current. Is not essential to the establishment of Therefore, according to the present embodiment, not only is the labor required for setting the voltage pulse greatly reduced, but also in the case where a specific change of the motor or a voltage fluctuation of the battery occurs, the accurate electrical angle can be stabilized. Can be detected.

Further, according to the present embodiment, the voltage application is stopped when the current rises due to the influence of the magnetic saturation, so that the voltage application time can be minimized, and the polarity determination can be performed in a short time. It can be performed. As a result,
The occurrence of abnormal noise during voltage application can also be reduced.

In the above embodiment, the current rise times T1 and T2 are used. However, the parameter for determining the polarity is based on any rate of change in current flowing when a voltage is applied to the coil. It may be. For example, after a voltage is applied, the current changes linearly (see FIG. 1).
A straight line approximation of the times t11 to t15 in the 0 pulse pattern # 1) is used, and the time until the difference between the straight line and the detected current value becomes equal to or more than a predetermined value is used as a parameter.

Also, various methods can be used to calculate the current change rate in the forward current rise time measurement processing (FIG. 9) and the reverse current rise time measurement processing (FIG. 12). For example, instead of stopping the application of the voltage when the current rises as in the present embodiment, the voltage is applied for a certain period in which magnetic saturation can be sufficiently generated in both the forward and reverse directions. After detecting all the behaviors of the current, the time required for the current to rise may be determined. In the method of the present embodiment, the current rise time is always an integral multiple of the sampling time Ts, and therefore includes an error of a maximum Ts / 2. By doing so, the current rise time can be obtained with high accuracy.

The voltage application period is the same as in this embodiment,
The calculation method of the current change rate may be changed. For example, in addition to the current value A1 detected at a certain time and the current value A0 detected at the immediately preceding sampling time, a current value A2 detected at the immediately preceding sampling time is used to determine three points. The current change rate may be obtained by obtaining a passing quadratic curve (a quadratic function with respect to time t: A = at2 + bt + c). When the current is changing linearly, the coefficient a of the square of the time t has a value of 0 or a very small value. On the other hand, if it increases rapidly, the coefficient also increases. Also, by differentiating the quadratic curve obtained in this way, the instantaneous current change rate can be calculated, so that the current rise time can be accurately obtained.

(3) Electrical Angle Detecting Process According to Second Embodiment Next, an electrical angle detecting process according to a second embodiment of the present invention will be described with reference to FIG. The configuration of the electrical angle detecting device in this embodiment is the same as the configuration (FIGS. 1 to 4) in the above-described embodiment.

The electric angle detection routine shown in FIG. 14 is a routine periodically executed by the CPU 120 together with various routines for controlling the operation of the motor 40. When the electric angle detection routine is started, the CPU 120
Performs a pulse pattern determination process (step S10).
0), a forward current rise time (T1) measurement process is performed (step S200). These processing contents are the same as the processing contents described in the above-described embodiment (FIGS. 6 and 9).

In the next step, the CPU 120
It is determined whether the forward current rise time T1 is longer than a predetermined time T0 (step S50).
0). The predetermined time T0 is a time required from the start of voltage application to the coil to the occurrence of magnetic saturation when the electrical angle is in a specific state. For example, when the electrical angle is π / 2, the time required for applying a voltage to the U phase and causing the current change rate to exceed the predetermined value ε can be set as the predetermined time T0. By setting the predetermined time T0 in this manner, the polarity can be determined as follows based on the magnitude relationship between the forward current rise time T1 and the predetermined time T0.

When the time T1 is longer than the predetermined time T0 and the electric angle φ belongs to the divisions 1 to 3 or the divisions 1 ′ to 3 ′, the electric angle θ = φ2, and the electric angle φ becomes the division 4 6 or 4 ′ to 6 ′, the electrical angle θ = φ1 (step S505). On the other hand, when the time T1 is equal to or shorter than the time T0, the electrical angle θ is obtained from the opposite relationship. When the electric angle φ belongs to the sections 1 to 3 or the sections 1 ′ to 3 ′, the electric angle θ = φ1, and when the electric angle φ belongs to the section 4 to 6 or the sections 4 ′ to 6 ′, Is the electrical angle θ = φ2 (step S51).
0).

The reason for making such a determination will be described with reference to FIG. The forward current rise time T1 is the shortest when the electrical angle is 0, and increases as the electrical angle increases. In FIG. 13, when the N pole exists in the hatched area (sections 1 to 3 and sections 4 ′ to 6 ′), the forward current rise time T1 is maximized because the magnetic pole is higher than the U-phase coil. This is the case where it is at the position of π / 2. In this embodiment, the forward current rise time in this state is set to the predetermined time T0. Therefore, the time T
When 1 is smaller than the predetermined time T0, it can be determined that the electrical angle exists in the sections 1 to 3 or the sections 4 'to 6'. When a voltage is applied to the V phase and the W phase, a predetermined time T
Since 0 can be set to substantially the same value, the polarity can be similarly determined. Therefore, steps S505 and S51
As described above, the electrical angle θ can be specified.

According to the present embodiment, similarly to the above-described embodiments, the electrical angle can be detected without requiring fine adjustment of the voltage pulse applied to the coil winding. Further, according to the present embodiment, since it is not necessary to measure the electrical angle reverse current rise time, the electrical angle can be detected in a shorter time than in the above-described embodiment.

In the present embodiment, as a result of determining the phase to which the voltage is applied by the pulse pattern determination processing (step S100), for example, when the voltage is applied to the U phase, the electrical angles are classified into categories 1, 1 ', and 6. , 6 ′. In other words, the angle between the U phase and the magnetic pole is 30
It will be in the range of degrees (π / 6). Therefore, the predetermined time T0 may be set to a forward current rise time in an electrical angle range of π / 6 to π / 2. By using in combination with the means for judging the section where the electric angle exists, the polarity judgment in the above embodiment can be performed stably without being affected by the change in the characteristics of the motor due to the passage of time or the temperature. Corners can be detected.

Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and various embodiments can be made without departing from the gist of the present invention. For example, the calculation of the electrical angle in the pulse pattern determination process (step S100) of the electrical angle detection routine is performed without determining the section in which the electrical angle exists, from 0 to π or from π to 2π.
The electric angle φ may be calculated in the range of (1).

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a motor control device 10.

FIG. 2 is an explanatory diagram showing a schematic configuration of a three-phase synchronous motor 40.

FIG. 3 shows a stator 30 and a rotor 50 of a three-phase synchronous motor 40;
FIG. 6 is an end view showing the relationship with FIG.

FIG. 4 is an equivalent circuit diagram of the three-phase synchronous motor 40.

FIG. 5 is a flowchart of an electrical angle detection routine in the first embodiment.

FIG. 6 is a flowchart of a pulse pattern determination process in the first embodiment.

FIG. 7 is an explanatory diagram for explaining an existing section of an electrical angle.

FIG. 8 is an explanatory diagram illustrating another existing section of the electrical angle.

FIG. 9 is a flowchart of a forward current rise time measurement process.

FIG. 10 is a graph showing a state of a voltage and a current of a coil winding.

FIG. 11 is a graph showing a relationship between an external magnetic field applied to a tooth 22 and an internal magnetic flux density.

FIG. 12 is a flowchart of a reverse current rise time measurement process.

FIG. 13 is an explanatory diagram showing a relationship between an electrical angle existence category and polarity determination.

FIG. 14 is a flowchart of an electrical angle detection routine in a second embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Motor control apparatus 20 ... Plate stator 22 ... Teeth 24 ... Slot 30 ... Stator 32 ... Coil 34 ... Bolt 36 ... Bolt hole 40 ... Three-phase synchronous motor 50 ... Rotor 51, 52, 53, 54 ... Permanent Magnet 55 Rotating shaft 57 Plate rotor 60 Case 61, 62 Bearing 71, 72, 73, 74 Salient pole 100 Control ECU 102, 103, 104 Current sensor 106, 107, 108 Filter 112 , 113, 114 ADC 116 input port 118 output port 120 CPU 122 ROM 124 RAM 126 clock 130 voltage applying unit

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H02P 5/00 H02P 5/28-5/44 H02P 6/00-6/24 H02P 7/00-7 / 01 H02P 7/36-7/66

Claims (2)

(57) [Claims] 1. An electric angle detecting device for a synchronous motor, wherein a polyphase alternating current flows through a winding, and a rotor rotates by an interaction between a magnetic field generated by the winding and a magnetic field generated by a permanent magnet. Voltage applying means for applying a voltage of the following; current detecting means for detecting a behavior of a current such as a rate of change of a current flowing according to the applied voltage; and a behavior of a current flowing according to the applied voltage.
An electrical angle calculating means for calculating an electrical angle within a section of electrical angles of 0 to π or π to 2π; and an electrical angle of 0 to π or π to 2π based on the variation of the rate of change of the detected current. Section specifying means for specifying which section belongs to , wherein the voltage applying means applies a predetermined voltage in one direction of the winding for a first predetermined period.
Forward voltage applying means for applying the predetermined voltage for a second predetermined period in a direction opposite to the one direction.
And a reverse voltage applying means for applying the same voltage as the voltage.
The section identification means detects the voltage from the start of voltage application by the forward voltage application means.
Required until the fluctuation value of the current change rate exceeds the specified value.
From the forward current rise time and the start of voltage application by the reverse voltage application means.
Required until the fluctuation value of the current change rate exceeds the specified value.
Based on the magnitude relationship with the reverse current rise time
Electric angle belongs to any section from 0 to π or π to 2π
Electrical angle detection device, which is a means for specifying whether 2. The electrical angle detection device according to claim 1 , wherein the first predetermined period in the forward voltage applying means is a period until the detected current change rate exceeds a predetermined value. The second predetermined period in the reverse voltage applying means is a period until the detected fluctuation value of the current change rate exceeds a predetermined value or the second predetermined period. An electrical angle detection device which is a shorter one of the periods exceeding one predetermined period.