JPH06315291A - Magnetic flux position calculation method of induction motor and control method using it - Google Patents

Abstract

(57) [Summary] [Object] The present invention relates to a speed control of an induction motor, which enables high-accuracy speed control including near zero speed, and has a speed sensor-less vector control method having equivalent performance. To provide. This is achieved by providing a magnetic flux position calculator for measuring the leakage inductance of the electric motor and detecting the position of the magnetic flux inside the vector control device of the induction motor, and accurately estimating the position of the magnetic flux. [Effect] Since the position of the magnetic flux in the electric motor can be estimated without being affected by the primary resistance of the induction motor, vector control based on the magnetic flux position becomes possible, and highly accurate including near zero speed. Speed and torque control can be realized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for controlling the speed of an electric motor by means of a power converter such as an inverter, and more particularly to a method for controlling the speed of an AC electric motor capable of performing highly accurate control over an extremely low speed to a wide speed range.

[0002]

2. Description of the Related Art At present, a slip frequency control type vector control method, which is widely used for driving steel rolling mills, FA servo drives, etc., uses an inverter output according to the sum of a slip frequency command value and an actual rotation speed. Since the frequency is controlled, a speed sensor attached to an electric motor is indispensable, and the application is restricted accordingly. For this reason, some high-accuracy speed control methods that do not use speed sensors have been announced (for example, 1993 S. 9th Annual Meeting of the Institute of Electrical Engineers of Japan, Symposium S. 9 "Current status and problems of applying induction motor speed sensorless vector control"). However, both methods estimate the rotation speed based on the induced electromotive force associated with the rotation of the motor.Therefore, in the range where the rotation speed is close to zero and the electromotive force is small, the speed estimation accuracy is affected by the primary resistance drop. Is deteriorated and the control accuracy of speed and torque is insufficient.

[0003]

In addition to the above, as a method of controlling without being affected by the primary resistance drop, there is provided a method of providing a search coil inside the motor, a third method of controlling the voltage and current of the motor.
There is a method of detecting the next harmonic component or a method of detecting the slot harmonic voltage of the electric motor. However, both methods are similar to the above-mentioned method in that the electromotive force based on the change in the primary interlinkage magnetic flux due to the rotation of the electric motor is detected, and the electromotive force is small near zero speed. The S / N ratio decreases with respect to the harmonic ripple from the superimposed inverter, and similarly high-precision control is difficult. Further, both methods have a drawback that the electric motor structure is specialized.

An object of the present invention is to solve the above problems and to provide a vector controller without a speed sensor, which enables high-accuracy speed control including near zero speed and has the same performance as that with a speed sensor. is there.

[0005]

In order to achieve the above object, a sine wave AC voltage is superposed on an output voltage command value of an inverter, and a motor current component flowing in response to the sine wave AC voltage is detected. Measure the leakage inductance of the motor winding. Based on the phenomenon that the inductance value changes due to the positional relationship between the winding and the motor magnetic flux, the magnetic flux position (rotation angle) is estimated from the inductance value, and the inverter output voltage phase is controlled based on this estimated magnetic flux to control the motor current The excitation component and the torque component (corresponding to the secondary current) are controlled.

[0006]

Function: A magnetic flux is generated inside the motor according to the voltage / current of the motor. Therefore, magnetic saturation occurs in the iron core portion through which the magnetic flux passes (high saturation). The same applies to the teeth portion in which the primary winding is housed, and the portion located in the magnetic flux direction has a high degree of saturation. The leakage inductance of the primary winding changes under the influence of the magnetic saturation of the teeth. Therefore, as described above, an AC voltage different from the fundamental wave component is superimposed on the motor voltage, the inductance of the winding is measured from the relationship between the current flowing thereby and the AC voltage, and the change in this inductance causes the magnetic flux to change. The position (rotation angle) of is estimated. By controlling the inverter output voltage / current with this magnetic flux position as the coordinate reference, non-interference control (vector control) between the torque of the electric motor and the magnetic flux is performed.

As a result, since the interference of the torque with respect to the magnetic flux (flux fluctuation) can be prevented, the estimation accuracy of the slip frequency and the rotation speed can be improved, and high-accuracy speed control can be performed.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention applied to a vector control system without a speed sensor will be described with reference to FIG. In the figure, 1 is an inverter that outputs a voltage proportional to the voltage command v ₁ *, 2 is an induction motor, and 3 is a current command i _1d *, i _1q * and an output frequency command ω ₁ *
Voltage command calculator that outputs voltage commands v _1d *, v _1q *, 4 is a coordinate converter that calculates three-phase voltage command v ₁ * from v _1d *, v _1q *, 5 is pulse width modulation of v ₁ * A PWM signal generator that converts the signal into a signal and performs PWM control of the inverter output voltage, 6 is a current detector that detects a motor current, and 7 is a converter that converts the motor current into a rotating magnetic field coordinate. The exciting current i _1d and the torque current i _1q are converted. Current component detector to detect, 8 is torque current command i _1q * and i _1q
, A current controller that amplifies the difference between and outputs the frequency command ω ₁ *,
9 amplifies the difference between the exciting current commands i _id * and i _1d , and outputs v
A current controller for adding to _1d *, 10 is a phase calculator for integrating ω ₁ *, and outputs a phase reference signal θ *, 11 is a speed command ω _r
A speed command circuit that outputs *, 12 is a slip frequency calculator that estimates the slip frequency ω _s based on i _1q , and 13 is ω
_A speed regulator that outputs i _1q * according to the difference between _r * and the estimated speed value ω _r ^ to control the speed. 14 adds sine wave signals v _1d ″, v _1q ″ to v _1d *, v _1q * Then, it is a magnetic flux position calculator that estimates the magnetic flux position φ _{1 of the} electric motor based on the component of i _1d flowing thereby.

Next, the operation of this control system will be described. For details of the parts 1 to 13 except for part number 14, see Okuyama, Fujimoto et al. In the Institute of Electrical Engineers of Japan “Velocity / voltage sensorless vector control method for induction motors”, D.
7, 2, pp 191-198 (1962), an outline is given here.

The system is roughly divided into three parts. The first part is an output voltage controller, which is composed of a voltage command calculator 3, a coordinate converter 4, and a pulse width modulator 5. FIG. 2 shows a vector diagram of the motor voltage v ₁ and the current i ₁ . Here, the dq axes are Cartesian coordinates that rotate at the synchronous speed ω ₁ . v ₁ is the induced electromotive force e ₁ ′ and the leakage impedance drop (r ₁ i ₁ , ω ₁ (l ₁ + l ₂ ′) as shown in the figure.
It is given by the sum of i ₁ ). Therefore, v command value of each axis component of ₁ v _1d *, v _1q * according the number 1, 'command value e ₁ of' e ₁ *
(= Ω ₁ * (M / L ₂ ) Φ _2d *) and the leakage impedance drop estimated value.

[0011]

[Equation 1]

Further, in the coordinate converter 4, v _1d *,
A three-phase voltage command value v ₁ * is calculated from v _1q *. Since the respective phase signals differ only by 120 ° in phase from each other, if only the u-phase voltage command v _u * is shown, then the equation 2 is obtained.

[0013]

[Equation 2]

Further, in the pulse width modulation 5, v ₁ * is converted into a pulse width modulation signal, and the output voltage of the inverter 1 is controlled accordingly. In this way, the fundamental wave component instantaneous value v _{1 of the} inverter output voltage is controlled in proportion to v ₁ *, and v ₁ is controlled based on v _1d *, v _1q *, and θ *. At this time, if I match the leakage impedance drop estimate the actual value of the above, 'the actual value of the command value e _1' e ₁ matches *, its orientation coincides with the q-axis. Under this condition, the phase reference θ * is the actual magnetic flux vector (e
₍ Orthogonal to ₁ ') and the rotation angle θ from the stator u phase axis,
θ * is equivalent to the rotation angle θ of the magnetic flux.

The second part is a current controller, which is composed of a current detector 6, a current component detector 7, and two current regulators 8 and 9.

As described above, under the condition that the direction of e ₁ ′ coincides with the q-axis, the current component detector 7 calculates i _1d and i _{1q as} the exciting current i ₀ ′ and the secondary current, respectively. It matches the current i ₂ ′.

[0017]

[Equation 3]

Therefore, by correcting v _1d * according to the control deviation of i _1d by the current regulator 9, i ₀ ′ is i
Controlled to match _1d *. At this time, the magnetic flux of the motor is controlled in proportion to i _1d *. In addition, the current regulator 8
_Therefore , depending on the control deviation of i _1q , ω ₁ * and e ₁ ′ * (=
ω ₁ * (M / L ₂ ) Φ _2d *) is controlled so that i ₂ ′ is controlled to match i _1q *. At this time, the electric motor generated torque τ _e is expressed by _Equation 4, and is proportional to i _1q *.

[0019]

[Equation 4]

The third part is a speed controller, which is composed of a speed command circuit 11, a slip frequency calculator 12, and a speed controller 13. In the calculator 12, the estimated value ω _s ^ of the slip frequency is calculated according to Equation 5.

[0021]

[Equation 5]

Next, ω _s ^ is subtracted from ω ₁ * to obtain the estimated speed value ω _r ^, and the speed command value ω _{r is} calculated in the speed controller 13.
I _1q * is calculated based on the difference between * and ω _r ^. In this way, as a result of controlling i _1q and torque τ _e , ω _r
Speed control is performed so that ^ matches ω _r *.

The above is the basic operation of the vector control without speed sensor. By the way, according to this method, speed control can be performed with sufficient accuracy in an operating frequency range of 1 Hz or higher.
However, in the low frequency range of 1 Hz or less, the rotational speed and torque control accuracy deteriorates. It is considered that this problem is mainly caused by the fluctuation of the motor primary resistance r ₁ . That is, when r ₁ fluctuates due to a change in the temperature of the electric motor, the estimated primary resistance drop (r ₁ * i ₁ *) used in Equation ₁ does not match the actual primary resistance drop (r ₁ i ₁ ). . At this time,
'actual value of the command value e _1' e ₁ varied from *, the orientation of e ₁ 'is not match the q-axis. Under the condition that the frequency is low and e ₁ ′ is small, the proportion of the primary resistance drop in the voltage v ₁ increases, and this tendency becomes remarkable. In this way, in low frequency operation, e ₁ ′ has a “deviation” from the q axis due to primary resistance fluctuation (estimation error of primary resistance drop). At this time, the phase reference θ * does not match the actual magnetic flux phase θ, vector control becomes incomplete, and the torque τ _e is no longer proportional to i _1q *. Further, since the magnetic flux Φ _2d is changed in relation to the torque (i _1q ), an estimation error also occurs in ω _s ^ calculated according to Equation 5, and an error also occurs in ω _r ^. As a result, the speed and torque control accuracy deteriorates.

The above problems are common to the vector control without a speed sensor, and various methods have been proposed as countermeasures as described above, but there is currently no drastic method. Therefore, in the present invention, the magnetic flux position calculator 14 is added to solve the above problem. The magnetic flux position calculator 1 is shown in FIG.
An outline of the calculation contents of 4 is shown. In the figure, 31 is a signal generator that outputs a two-phase sine wave signal (sin ωt, cos ωt), 32 is a signal (sin ωt) input, and mode 1, 2, 3
According to (1 / √2) sinωt, (1 / √2) sinωt, sin
A switch circuit for outputting a signal of ωt, 33 is a mode 1,
Depending on 2, 3 (1 / √2) sinωt, − (1 / √2) sinω
Switch circuits that output signals t and 0, 33 and 34 are multipliers that multiply the current i _1d by the signals sin ωt and cos ωt, 36 and 37 are integrators that integrate the outputs of the multipliers 34 and 35, and 38 is an integral An inductance calculator that measures the inductance values Lσ ₁ , Lσ ₂ and Lσ ₃ in each mode based on the output values of the devices 36 and 37, and 39 is a calculator that calculates the position angle φ _{1 of the} magnetic flux of the motor based on each Lσ Is.

Next, the contents of calculation will be described. First, the basic principle of estimating the magnetic flux position angle φ ₁ will be described. FIG. 4 shows a model of the induction motor. Assuming that the magnetic flux exists in the direction of Φ in the electric motor, magnetic saturation occurs (high saturation) in the iron core portion located in the Φ direction. The same applies to the teeth portion in which the primary winding is housed, and the portion located in the Φ direction has a high degree of saturation. The leakage inductance of the primary winding changes due to the influence of magnetic saturation of the teeth. For example, as shown in the drawing, the leakage inductance of the winding A located in the Φ direction is smaller than that of the winding B orthogonal to the Φ direction. FIG. 5 shows the result of this measurement, and shows the change of the leakage inductance value of each winding with respect to the exciting current (flux amount). As shown in the figure, it was confirmed by experiments that the inductance value significantly changes depending on the positional relationship between the magnetic flux and the winding near the rated exciting current (3 A).

Therefore, the magnetic flux position (direction) can be estimated conversely by detecting the change in the inductance.
By controlling the output voltage / current of the inverter based on the estimated magnetic flux position as the coordinate reference, it is possible to perform vector control with high accuracy without being affected by the above-mentioned change in primary resistance. The above is the basic principle of magnetic flux position estimation.

Next, the principle of measuring the inductance Lσ will be described. Now, a sine wave voltage v (= sinωt) having a frequency different from the fundamental wave is applied to the electric motor, and the alternating current i flowing by this is observed. Under the condition that the angular frequency ω of v is sufficiently higher than the reciprocal of the secondary time constant T ₂ of the electric motor, the transfer function of the winding current / voltage can be approximated by a first-order lag system, so i
Indicated by.

[0028]

[Equation 6]

Fourier transform of i with respect to v is performed to obtain a component synchronized with v and a 90 ° phase difference component.
When Lσ is calculated on the assumption that it is equal to the first term and the second term on the right-hand side of

[0030]

[Equation 7]

As described above, Lσ can be measured based on v and i.

Next, the basic principle of magnetic flux position estimation and the operation of the calculator 14 will be described. Now, as shown in FIG. 6, the angle between the direction of the magnetic flux Φ and the direction of the magnetomotive force of the winding C to which the AC voltage for measuring the inductance is applied is defined as φ. L
Since σ is minimum at φ / 2, 3π / 2, and maximum at 0, π, L _σ changes as a function of 2φ. Therefore, L _σ can be set as in Expression 8.

[0033]

[Equation 8]

Where φ = φ ₁ + π / 4, and φ = φ ₁
An alternating voltage is applied to each −π / 4 winding in order, and L _σ is measured as described above. _Let each L _σ be L _σ1 and L _σ2 ,

[0035]

[Equation 9]

[0036]

[Equation 10]

From equations 9 and 10

[0038]

[Equation 11]

Further, by applying an AC voltage to the winding of φ = φ ₁ and measuring L _σ3 ,

[0040]

[Equation 12]

From equations 9 and 12,

[0042]

[Equation 13]

That is, φ = φ ₁ + π / 4, φ ₁ −π /
4, the 3-point measurement of phi _1, it is possible to obtain the phi _1, can be estimated flux position.

The arithmetic unit 14 operates based on the above-mentioned estimation principle. The operation will be described below with reference to the vector diagrams of FIGS. 3 and 7. For vector control, it is ideal that the directions of the d-axis and the magnetic flux Φ coincide with each other, but here, assuming that they do not coincide with each other, the angle difference φ ₁ is assumed. Less than,
Each case of modes 1, 2 and 3 will be described in order.

[Mode 1] Sine wave signal ((1 / √2) sinω
t) is, as v _1d "through the switch circuit 32, v _1d
7 is also added to v _1q * as well as v _1q ″ via the switch circuit 33. This state is in the direction of mode 1 (45 ° with respect to the d axis in the vector diagram of FIG. 7). ) Corresponds to the application of the alternating voltage v to the winding having the direction of the magnetomotive force.
Flows. Even if i is observed from the d axis, i has the same phase as i, so i can be detected from i _1d . Therefore, in the multipliers 34 and 35, the integrators 36 and 37, and the inductance calculator 38, i _1d and the signal (sinωt, cosωt)
Based on the above, the calculation of _Equation 7 is performed to obtain L _σ1 .

L _σ1 is stored in the calculator 38.

The [mode 2] signal ((1 / √2) sinωt) is set as v _1d ″ through the switch circuit 32, and (− (1 / √2) sinωt) is set as v _1q through the switch circuit 33. ″ ”Is added to each voltage command value. In this state, v is applied in the direction of mode 2 in FIG. 7, and i flows in the same direction. Since i can be detected from i _1d similarly to the above, L _σ2 can be obtained by performing the same calculation as in mode 1. L _σ2 is similarly stored.

[Mode 3] The signal (sin ωt) is set to v _id ″, and v _1q ″ = 0, and added to each voltage command value. In this state, v in the direction of Mode 3 (d-axis) in FIG.
Is applied, and i flows in the same direction (d-axis). i can be detected as it is from i _1d , and L can be detected as described above.
_σ3 is _calculated and stored.

L _σ1 , measured as described above,
Based on L _σ2 and L _σ3 , the arithmetic unit 39 calculates
The magnetic flux position angle φ ₁ with respect to the d-axis is estimated by performing the calculations 1 and 13.

Next, the operation of the entire vector control system without a speed sensor to which the present invention is applied will be described with reference to FIG. The basic operation of the system is as described above. In order to solve the problem of accuracy deterioration in low speed operation, the magnetic flux position calculator 14 is added in the present invention. AC signals v _1d ″, v _1q ″ from the calculator 14
Is always added to v _1d *, v _1q * during operation, so that the current i _1d includes a current component related to v _1d ″, v _1q ″. Although i _1d includes a direct current component related to the fundamental wave component of the motor current, it is deleted in the calculation of L _σ in equation 7. Therefore, φ ₁ is measured irrespective of the operating condition, regardless of the magnitude of the rotation speed, and is not affected by the fluctuation of the primary resistance.

This φ ₁ corresponds to the “deviation angle” in the magnetic flux direction from the d-axis, as described above. Therefore, the arithmetic unit 10
Add a signal (proportional or integral value of φ ₁ ) according to φ ₁ to the input side or output side of to correct θ *. At this time, the phase reference value (corrected θ *) used for the coordinate converter 4 and the current component detector 7 matches the actual magnetic flux phase θ.

Thus, conventionally, even in the low speed operation range where θ * and θ do not match due to the variation of the primary resistance, θ
Since * = θ, vector control can be performed with high accuracy. As a result, the motor generated torque τ _e becomes proportional to i _1q * in the entire speed range including zero speed. Also, since there is no magnetic flux fluctuation associated with torque change, ω _s
^ Is estimated correctly, is estimated at the same time ω _r ^ is also a high degree of accuracy. From these, according to the present invention, highly accurate control of speed and torque can be performed in the entire range including zero speed.

FIG. 8 shows another embodiment of the present invention. The present invention is applied to a vector control system without a speed sensor, which is a system provided with an AC current control system that feedback-controls the instantaneous output current i ₁ of the inverter and controls it to follow the sine wave current command i ₁ *. Here is an example.

In the figure, 81 is a sine wave current command i ₁ *
Inverter that outputs a current proportional to the current, 82 is an induction motor, 83 is current commands i _1d *, i _1q *, and phase reference θ *
A coordinate converter for calculating a three-phase current command i ₁ * based on 8
4 is an AC current controller that inputs the difference between i ₁ * and the current i ₁ and outputs a voltage command v ₁ *; 85 is a PWM that converts v ₁ * into a pulse width modulation signal and PWM-controls the inverter output voltage Signal generator, 86 is a current detector for detecting electric motor current, 8
Reference _{numeral 7} is a current component detector that detects a torque current i _1q by converting i ₁ into a rotating magnetic field coordinate by using a magnetic flux obtained by integrating the motor voltage v ₁ as a phase reference, and 88 is a torque current command i _1q * and i _1q. The current regulator 89 outputs an estimated rotational speed value ω _r ^ according to the difference between the two, and 89 is a slip frequency estimated value ω _s based on i _1q *.
A slip frequency calculator for calculating ^, 90 is ω _r ^ and ω _s ^
Integrating the frequency command omega ₁ * obtained by adding a phase calculator for outputting a phase reference signal theta *, 91 speed command circuit for outputting a speed command omega _r *, 92 is omega _r * and omega _r ^ A speed regulator that outputs i _1q * according to the difference between the two, and controls the speed, 93 detects the electric motor voltage v ₁ and converts it into a rotating magnetic field coordinate based on θ * and detects the d-axis voltage v _1d . A voltage component detector 94 adds a sine wave signal i _1d ″, i _1q ″ to i _1d *, i _1q *, and calculates a magnetic flux position angle φ ₁ based on the component of v _1d generated thereby It is a vessel.

Next, the operation of this control system will be described. This system can also be roughly divided into three parts. The first part is an output current controller, which is composed of a coordinate converter 83, a current controller 84, a pulse width modulator 85, and a current detector 86. In the coordinate converter 83,
A three-phase current command value i ₁ * is calculated from the d-q axis current command values i _1d *, i _1q *. Since the respective phase commands only differ in phase by 120 ° from each other, if only the u-phase current command value iu * is shown, then the formula 14 is obtained.

[0056]

[Equation 14]

In the current controller 84, the output voltage command value v ₁ * is calculated according to the difference between i ₁ * and i ₁ . Furthermore, P
In the WM modulator 85, v ₁ * is converted into a pulse width modulation signal, and the output voltage of the inverter 1 is controlled accordingly. As a result, since i ₁ is controlled in proportion to i ₁ *, i ₁ is controlled based on i _1d *, i _1q *, and θ *.

The second part is a speed estimator, which is a current component detector 87, a current regulator 88, and a slip frequency calculator 8
9 and a phase calculator 90. Current component detector 8
In FIG. 7, first, the motor magnetic flux Φ is detected according to the equation (15).

[0059]

[Equation 15]

Φ is divided by the amplitude value | Φ | to calculate a sinusoidal magnetic flux phase signal (sin θ, cos θ) having a constant amplitude. Based on the signal, according to equation 3 (replace θ * with θ), i
_1q is calculated. In the current regulator 88, ω _r ^ is calculated based on the difference between i _1q * and i _1q . That is, as described above, under the condition that i ₁ is controlled according to i _1d *, i _1q *,
The difference between i _1q and i _1q * is caused by the fact that θ *, which will be described later, does not match the magnetic flux phase θ, and ω ₁ * is controlled by the current regulator 88 to correct this. In this way, θ * =
Under the condition that θ 1 is satisfied, the vector control is correctly performed and the magnetic flux of the motor is kept at a predetermined value. As a result, in the slip frequency calculator 89, _Equation 5 (replace i _1q with i _1q *)
Therefore, ω _s ^ is correctly estimated. In addition, ω _r ^ (= ω ₁ *
-Ω _s ^) is correctly estimated. In the phase calculator 90, ω ₁ * is integrated to obtain θ *, which is used as the phase reference of the coordinate converter 83.

The third part is a speed control section, which is composed of a speed command circuit 91 and a speed adjuster 92. I _1q * is calculated based on the difference between the speed command values ω _r * and ω _r ^.

Since the torque τ _e is controlled according to "Equation 4" according to i _1q *, the speed control is performed so that ω _r ^ matches ω _r *. The above is the basic operation of the vector control without the speed sensor. However, even in this case, the control accuracy is deteriorated due to the fluctuation of the primary resistance in the low frequency operation. This is because when the magnetic flux Φ ^ is calculated in the current component detector 87, r ₁ is used as a calculation constant as shown in Expression 15.
This is because an error occurs in Φ ^ when this does not match the actual r ₁ . A detection error also occurs in i _1q due to the phase shift between Φ ^ and Φ, and θ * = θ 2 is not established, resulting in incomplete vector control. As a result, the control accuracy of speed and torque deteriorates as in the previous embodiment.

Therefore, a voltage component detector 93 and a magnetic flux position calculator 94 are added to solve this problem. FIG. 9 shows an outline of the calculation contents of the magnetic flux position calculator 94. In the figure, 101 is a signal generator that outputs a two-phase sine wave signal (sin ωt, cos ωt), 102 is a signal sin ωt,
Depending on modes 1, 2 and 3, (1 / √2) sinωt, (1 /
√2) A switch circuit that outputs signals of sinωt and sinωt, 103 is (1 / √2) si according to modes 1, 2, and 3.
nωt, − (1 / √2) sinωt, a switch circuit that outputs a signal of 0, 104 is a multiplier that multiplies the voltage v _1d and the signal cosωt, 105 is an integrator that integrates the output of the multiplier 104, and 106 is an integral An inductance calculator that measures the inductance values L _σ1 , L _σ2 , and L _σ3 in each mode based on the output value of the device 105, and 107 is a calculator that calculates the position angle φ _{1 of the} magnetic flux of the motor based on each L _σ. .

Next, the principle and contents of the magnetic flux position calculation will be described. This basic concept is the same as above. Further, the calculation contents of the arithmetic unit 107 are also the same as those of the arithmetic unit 39 of FIG. 3, so the contents until _obtaining L _σ1 , L _σ2 and L _σ3 will be described.

Now, a sine wave current i (= sinωt) having a frequency different from that of the fundamental wave is passed through the motor, and the AC voltage v generated from this is observed. Under the condition that the angular frequency ω of i is sufficiently higher than the reciprocal of the quadratic time constant T ₂ , the transfer function of voltage / current can be approximated by a first-order lead system, so v is expressed by Expression 16.

[0066]

[Equation 16]

On the other hand, the detected v is Fourier-transformed with i as a reference to obtain a component in phase with i and a 90 ° phase difference component,
Letting each be equal to the first term and the second term on the right-hand side of Equation 16, L _σ is Equation 17

[0068]

[Equation 17]

Here, | i | is the magnitude of the current and is a preset amount. As described above, the difference from the previous embodiment is that the AC voltage v is applied to the winding in the previous embodiment and L _σ is measured from the current i flowing thereby, whereas the AC current i is changed in the present embodiment. Is applied to the winding, and the voltage v generated from this is measured. Subsequent calculation, i.e., i is sequentially applied to the three windings of φ = φ ₁ + π / 4, φ ₁ −π / 4, φ ₁ , and from each v, L _σ1 , L _σ2 , L _σ3
And the calculation of φ ₁ is the same as in the previous embodiment. That is, in FIG. 9, the switch circuit 1
02 and 103 add predetermined i _1d ″ and i _1q ″ to i _1d * and i _1q * for each mode as described above, and the i component is superimposed on the motor current i ₁ . As a result, in addition to the fundamental wave component, modes 1, 2, 3 shown in FIG.
The alternating current i flows in the direction of and the alternating voltage v is generated in each direction accordingly. v can be detected from the d-axis voltage v _1d . Therefore, in the multiplier 104, the integrator 105, and the inductance calculator 106, v _1d and the signal (cosω
Based on t), the operation of _Equation 17 is performed, and L _σ1 , L
_{Calculate σ2} and L _σ3 . Then, the calculator 107 calculates φ ₁ based on each of the L _σ . Since the contents are the same as those of the arithmetic unit 38 of FIG. 3, the description thereof will be omitted.

Next, the operation of the entire system shown in FIG. 8 to which the present invention is applied will be described. The basic operation of the system is as described above. A magnetic flux position calculator 94 is added in order to solve the accuracy deterioration in the low speed operation, and this operation is as described above. From the viewpoint of operation, φ ₁ is measured regardless of the operating speed of the electric motor, irrespective of the magnitude of the rotation speed, and the influence of the primary resistance fluctuation. This φ ₁ corresponds to the “deviation angle” of the magnetic flux from the d-axis, as described above. Therefore, a signal corresponding to φ ₁ is added to the input of the current regulator 88 via a compensation element (coefficient unit or integrator) (negative polarity), and θ * is corrected so that φ ₁ becomes zero. . At this time, θ * comes to coincide with the actual magnetic flux phase θ.

As described above, conventionally, θ * = θ can be established even in the low speed range where θ * ≠ θ due to the variation of the primary resistance, and the vector control can be performed with high accuracy. As a result, according to the present invention, in this embodiment as well as in the previous embodiments, it is possible to perform high-precision control of speed and torque in the entire range including zero speed.

In the above embodiment, the magnetic flux position calculator is always operated during operation. However, even if the influence of the primary resistance fluctuation is large in the low speed range, the influence is small in the high speed range and there is almost no problem. Therefore, in the high speed range, even if the operation of the magnetic flux position calculator is stopped and the conventional control is performed, the same result as that of the above-described embodiment can be obtained as a matter of course.

Further, in the above embodiment, three L _σ are obtained during operation by obtaining three L _σ , and φ ₁ is calculated, but a single L _σ is obtained by measuring one point of φ. It is also possible to calculate and calculate φ ₁ . This is done as follows. In other words, before the actual operation of the electric motor during static operation (ω ₁ * =
In 0), i _1d * is set to a predetermined value, and a three-point measurement of φ is performed under the condition that a predetermined exciting current i _{1d is made} to flow and three L
_{Find σ} . If each is _expressed as L _σ10 , L _σ20 , and L _σ30 , then from _Eqs. 9 to 13, L _σm and a are given by Eq.
It is shown by the number 19.

[0074]

[Equation 18]

[0075]

[Formula 19]

Since L _σm , a does not change under the condition that the magnetic flux (exciting current) is constant, in actual operation, these values are calculated by the formula 1
By substituting in 2, φ ₁ can be obtained. That is,
AC signal is superimposed only on the d-axis voltage or current,
By measuring 1 point of φ, φ ₁ can be calculated according to the equation (20).

[0077]

[Equation 20]

During stationary operation, only L _σm is obtained by measuring φ at two points according to equation 18, and at actual operation, by measuring φ at two points, as described above, according to equations 19 and 20, It is also possible to obtain a and φ ₁ .

The AC signal used for the magnetic flux position calculation is not limited to a sine wave, and may be any other AC signal. This is because L _σ can be calculated in the same manner as described above by Fourier-transforming the motor current or voltage with the fundamental wave component of the AC signal as a reference.

In the above embodiment, the AC signal is superposed on the command signal of the inverter and L _σ is calculated based on the relational component contained in the motor current or voltage. However, these calculations are provided separately from the inverter. It can also be carried out independently using the above apparatus. The same effect can be obtained by sending φ ₁ obtained by the apparatus to the inverter and controlling the same.

[0081] In the above embodiment, although by changing the phase reference theta * by phi _1, a function of phi ₁ is not limited to phi _1, similarly used if phi ₁ and that there is a substantially proportional relationship The same effect can be obtained.

The embodiment using the magnetic flux position calculation method of the present invention is not limited to the speed sensorless vector control described above, and as another application, a power converter capable of arbitrarily controlling the output voltage and frequency. And an induction motor driven by the converter, the output frequency of the converter according to the added value of the rotation speed detection value of the motor and the slip frequency command value calculated by multiplying the torque current by a coefficient. In the motor control device for controlling the above, it is possible to easily change the coefficient of the slip frequency command calculation in accordance with the magnetic flux position signal obtained by the magnetic flux position calculation method described above.

According to this, even if the secondary resistance of the induction motor changes due to temperature or the like, the above coefficient is automatically corrected accordingly, so that there is an effect that the deterioration of the torque control accuracy can be prevented. Also, the same effect can be obtained by changing the output frequency according to the magnetic flux position signal.

[0084]

According to the present invention, since the position of the magnetic flux in the electric motor can be accurately estimated without being affected by the primary resistance of the induction motor, vector control based on the magnetic flux position becomes possible. , Highly accurate speed including near zero speed,
And torque control can be realized.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a vector control device without a speed sensor according to an embodiment of the present invention.

FIG. 2 is a vector diagram of motor voltage and current.

FIG. 3 is a configuration diagram of a magnetic flux position calculator in FIG.

FIG. 4 is a model of an induction motor.

FIG. 5 is a measurement result of leakage inductance according to the present invention.

FIG. 6 is a diagram showing a positional relationship between a magnetic flux of a motor and windings according to the present invention.

FIG. 7 is a vector diagram of a leakage inductance measurement mode according to the present invention.

FIG. 8 is a configuration diagram of a vector control device without a speed sensor according to another embodiment of the present invention.

9 is a configuration diagram of a magnetic flux position calculator in FIG.

[Explanation of symbols]

1 ... Inverter, 2 ... Induction motor, 3 ... Voltage command calculator, 4 ... Coordinate converter, 5 ... PWM signal generator, 6 ... Current detector, 7 ... Current component detector, 8 ... Current regulator, 9 ... Current regulator, 10 ... Phase calculator, 11 ... Speed command circuit, 1
2 ... Slip frequency calculator, 13 ... Speed controller, 14 ... Magnetic flux position calculator.

Claims (18)

[Claims] 1. An induction motor is superposed with an AC voltage having a frequency different from its driving voltage, the inductance of the motor is measured based on the motor current flowing at this time and the AC voltage, and the motor is based on this. A method for calculating the magnetic flux position of an induction motor, which is characterized by finding the position (angle) of the magnetic flux. 2. An induction motor is superposed with an alternating current having a frequency different from that of its drive current, and the inductance of the electric motor is measured based on the electric motor voltage and the alternating current generated at this time, and the electric motor is based on this. A method for calculating the magnetic flux position of an induction motor, which is characterized by finding the position (angle) of the magnetic flux. 3. A motor controller comprising a power converter capable of arbitrarily controlling an output voltage and frequency and an induction motor driven by the converter, wherein an output voltage (fundamental wave component) is set as an output voltage command value of the converter. ) Is superimposed, an inductance of the electric motor is measured based on the detected electric current value of the electric motor and the AC signal at this time, and the position (angle) of the electric motor magnetic flux is obtained based on the measured inductance. Flux position calculation method for induction motor. 4. A motor controller comprising a power converter capable of arbitrarily controlling an output current and frequency and an induction motor driven by the converter, wherein an output current command value of the converter is equal to an output current (fundamental wave). Component), an AC signal with a frequency different from that of the component) is superimposed, and the inductance of the motor is measured based on the detected value of the motor voltage and the AC signal at this time, and the position (angle) of the magnetic flux of the motor is determined based on this. Characteristic induction motor magnetic flux position calculation method. 5. The calculation method according to claim 3, wherein an AC signal is added to each voltage command value of the magnetic flux axis (d axis) on the control of the motor control device and the q axis orthogonal thereto, and the motor at this time is added. A method for calculating a magnetic flux position of an induction motor, which comprises measuring an inductance of the electric motor based on a converted value of a rotating magnetic field coordinate of an electric current and the AC signal and determining a position (angle) of a magnetic flux of the electric motor based on the measured inductance. 6. The calculation method according to claim 4, wherein an AC signal is added to each current command value of the magnetic flux axis (d axis) on the control of the electric motor control device and the q axis orthogonal thereto, and the electric motor at this time is added. A method for calculating a magnetic flux position of an induction motor, which comprises measuring the inductance of the electric motor based on a rotating magnetic field coordinate conversion value of the voltage and the AC signal, and determining the position (angle) of the electric motor magnetic flux based on the measured inductance. 7. The calculation method according to claim 3, wherein the voltage command values of the magnetic flux axis (d axis) on the control of the electric motor control device and the q axis orthogonal thereto are determined according to at least three modes. , AC signals are added by a predetermined value to generate alternating magnetomotive forces in three different directions inside the motor for each mode, and based on the rotating magnetic field coordinate conversion value of the motor current and the AC signal in each mode, Corresponding to the above, the method for calculating the magnetic flux position of the induction motor is characterized in that the position (angle) of the magnetic flux of the electric motor is obtained based on the measurement of three inductances. 8. The method according to claim 4, wherein the magnetic flux axis (d axis) on the control of the motor controller and the q-axis current command values orthogonal to the magnetic flux axis are set according to at least three modes. AC signals are added by a predetermined value to generate alternating magnetomotive forces in three different directions inside the motor for each mode. Based on the rotating magnetic field coordinate conversion value of the motor voltage at this time and the AC signal, each mode is generated. A method for calculating the magnetic flux position of an induction motor is characterized in that the position (angle) of the magnetic flux of the electric motor is obtained based on the three measured inductances. 9. The calculation method according to claim 7, wherein the voltage command value is set so that the directions of the alternating magnetomotive forces generated in the motor for each of the modes have an electrical angle difference of 45 °.
Alternatively, the magnetic flux position calculation method for an induction motor is characterized in that the magnitude of the AC signal to be added to the current command value is set. 10. A motor controller comprising an electric power converter capable of arbitrarily controlling an output voltage and a frequency and an induction motor driven by the converter, wherein an output voltage of the converter before actual operation of the electric motor. An AC signal of a frequency different from the output voltage (fundamental wave component) is superimposed on the command value or the output current command value, and the inductance of the motor is detected based on the detected value of the motor current or motor voltage and the AC signal. Is measured, and its characteristic amount (average value and / or fluctuation range) is calculated and stored. During actual operation, the inductance is measured in the same manner, and the magnetic flux of the motor is calculated based on the inductance value and the characteristic amount. The method for calculating the magnetic flux position of an induction motor is characterized by finding the position (angle) of 11. The calculation method according to claim 10, wherein before the actual operation of the electric motor, a magnetic flux axis (d axis) for control of the electric motor control device and each voltage command value or current command of a q axis orthogonal thereto. The AC signal is added to the value, and the inductance of the motor is measured based on the rotating magnetic field coordinate conversion value of the motor current or the motor voltage at this time and the AC signal, and its characteristic amount (average value and / or fluctuation range) is measured. Is calculated and stored, and during actual operation, the inductance is similarly measured, and the position (angle) of the magnetic flux of the motor is obtained based on the inductance value and the characteristic amount. Position calculation method. 12. The magnetic flux position calculation method according to claim 10, wherein before the actual operation of the electric motor, a magnetic flux axis (d axis) for control of the electric motor control device and each voltage command value of the q axis orthogonal thereto or An alternating current signal is superposed on each current command value by a predetermined value according to each of the three or two modes, and an alternating magnetomotive force is generated in each of the modes in three or two different directions inside the motor to generate an alternating magnetomotive force. Corresponding to 3
One or two inductances are measured, and the characteristic amount (both the average value and the fluctuation range, or one of them) is calculated and stored, and in actual operation, similarly, according to one or two modes, A method for calculating a magnetic flux position of an induction motor, wherein one or two inductances are measured, and the position (angle) of the magnetic flux of the motor is obtained based on the inductance value and the characteristic amount. 13. The frequency or phase of the output voltage / current of the power converter is obtained according to the magnetic flux position signal obtained by the magnetic flux position calculation method according to any one of claims 3 to 12, and based on this. A method for controlling an induction motor, comprising controlling the induction motor. 14. A coordinate converter for calculating an AC command value of an output voltage / current of the power converter according to a magnetic flux position signal obtained by the magnetic flux position calculating method according to any one of claims 3 to 12. A method for controlling an induction motor, comprising: controlling an induction motor by changing a phase reference value. 15. The control method according to claim 13 or 14, wherein the control method is performed only when the output frequency of the converter is equal to or lower than a predetermined value. 16. An electric power converter capable of arbitrarily controlling an output voltage and a frequency, and an induction motor driven by the converter, which is responsive to an added value of a rotation speed detection value of the electric motor and a slip frequency command value. In the motor controller for controlling the output frequency of the converter, the slip frequency command value or the frequency command value is set in accordance with the magnetic flux position signal obtained by the magnetic flux position calculation method according to any one of claims 1 to 12. A method for controlling an induction motor, which is modified. 17. A slip calculated by a power converter capable of arbitrarily controlling an output voltage and a frequency and an induction motor driven by the converter, the slip calculated by multiplying a detected rotational speed of the electric motor and a torque current by a coefficient. In a motor control device that controls the output frequency of the converter according to the added value with the frequency command value, according to the magnetic flux position signal obtained by the magnetic flux position calculation method according to any one of claims 1 to 12, A method of controlling an induction motor, characterized in that a coefficient of the slip frequency command calculation is changed. 18. An induction motor is generated by superimposing an AC voltage having a frequency different from its drive voltage, and based on the motor current and the AC voltage flowing at this time, a difference in the saturation state of the iron core of the motor. A method for calculating the magnetic flux position of an induction motor, which detects a physical phenomenon and obtains the position (angle) of the magnetic flux of the motor based on the detected physical phenomenon.