JPH01110091A - Variable speed controller for induction motor - Google Patents

Abstract

PURPOSE:To control the speed of an induction motor with high performance even at a low speed by obtaining a magnetic flux phase from a current model and a motor voltage, and obtaining a rotor speed indirectly with the difference at zero value. CONSTITUTION:The component u1delta (reactive component of a voltage) of a motor primary voltage vector V1 perpendicularly crossing a current vector is calculated by a vector rotor 30 from the voltage vector V1 being the output of a 3-phase/2-phase converter 21. A current model 10 calculates the phase phi1 of a magnetic flux. A speed presuming circuit 32 calculates a magnetizing current, the command values i'M, i'T of a torque current, a magnetic flux command value psi'2, the angular speed omega1 of the magnetic flux and a motor constant, calculates to so adjust, as predetermined, by an inner adjustor, that the deviation from actual reactive voltage value u1delta given from the rotor 30 becomes zero, and outputs a presumed speed value omegaV.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a high-performance variable speed control device for an induction motor that eliminates the need for a device that directly detects speed.

[Conventional technology]

As an example of a high-performance induction motor variable speed device that does not require a device that directly detects the speed, the applicant has developed a PWM motor as shown in Figure 5.
Inverter l is used as a power source, and vector 1fil is used for its control.
An application has been filed for a variable speed control device that applies the i# principle (Japanese Patent Application No. 60-21262: hereinafter simply referred to as the applied device) 0 The principle of vector control itself has already been published in many documents, etc. Although it is publicly known (for example, "Fuji Jiho")
@Volume 53, No. 9, P640-648, see section ``Transvector control of alternating current machine''. ), will be briefly explained here.

Vector control of an induction machine considers the current, voltage, etc. of the motor as vector quantities, and converts these R, which are AC quantities when observed from the stator windings, into DC quantities when observed from the rotating magnetic field of the motor. , this is separated into a component parallel to the magnetic field and a component perpendicular to the magnetic field, and each component is controlled independently.

FIG. 6 shows the primary current vector 11 of the induction machine at a fixed coordinate α
axis, component ia on the β axis (coordinate system taken on the stator winding)
, iβ and rotational coordinates M axis, T axis (axis taken on magnetic flux = i
This shows a state where the image is separated into components IM t IT on a coordinate system in which the M-axis and the T-axis are perpendicular to the M-axis. That is, the figure shows the relationship between six quantities on the stator coordinates (α-β) and the rotating coordinates (MT). Note that although vector quantities are shown with arrows, the distinction will not be made hereinafter unless particularly necessary.

Now, a vector control device without a speed detector will be explained with reference to FIG.

@ In diagram 5, the primary current of the induction motor 2 is 3-phase -2
The phase converter 12 converts it into two phase quantities ia and iβ. Further, this quantity is coordinate-transformed by a vector rotator (VD) 11 into a rotational coordinate (MT coordinate system) quantity IM*IT.

At this time, the M axis, that is, the magnetic flux axis, is determined by the phase φl of the magnetic flux calculated by a siege model type magnetic flux calculator (also simply referred to as a current model) 10, which will be described later. This coordinate transformation is performed using the following equation.

It is well known that if the primary current 11 is separated into IM s IT in this way, IM becomes the component that creates magnetic flux (magnetization 1! current) and IT becomes the component that creates torque (torque current). .

The magnetizing current command, +, is given as a result of passing the output W2 of the magnetic flux command calculator 4 through the differentiating circuit 3.

In the case of constant magnetic flux control, the calculator 4 gives a constant F2, and in the case of speed-dependent field weakening control in a high-speed region, the calculator 4 gives a constant F2, which decreases as the speed increases.
give. The differentiating circuit 3 performs the calculation of the following equation.

The magnetizing current command - is compared with - converted from the primary current by the vector rotator 11 at the addition point 14, and this deviation is amplified by the PI (proportional integral) l1m unit 6, and the primary voltage vector of the motor is Give the M-axis component vM of the command vl. The speed is calculated as the output of the regulator 25 in the regulation loop 200, the principle of which will be explained later.

On the other hand, the command value N given from the speed setting device 100 is
At the addition point 13, the estimated speed value ω7 which is the output of the regulator 25
This deviation is amplified by the PI regulator 5 and becomes the torque current command ET. This 1T4I is compared with ly produced by the vector rotator 11 at the addition point 15, and this deviation is amplified at the PIg moderator 7→extension, and the T-axis component v of the -th voltage vector command vl
It becomes T. This VM, Ver is the coordinate conversion circuit 8
The magnetic flux phase φIIC calculated by the current model 10 is converted into a stator coordinate quantity as shown in the following equation.

Primary voltage command V converted into fixed fawn S quantity.

The pulse Vβ is converted into an inverter pulse by the pulse generating circuit 9, and is applied to the PWM inverter 1 to be supplied to the induction motor 2. Here, the phase φ□ of the magnetic flux used when converting the coordinates of the voltage and voltage of the motor is calculated by the following equation using the *m model IO.

φ knee fω1d / (ω2 + ω, ) dt Old... (3) (a) t (1) From the formula, the phase φ1 of the magnetic flux is IM tiT' and the speed N is human power, and the motor constant (
It can be seen that the calculation can be performed if R2°M%) is known.

Note that in the case of constant magnetic flux control or when the magnetic flux changes slowly, it is also possible to use TtTzxQ in equation (4). The slip angular velocity calculator 101 executes the calculation of equation (4), and the integrator 102 executes the calculation of equation (3).

The principle for controlling the speed of an induction machine without a speed detector is that the phase φ of the magnetic flux of the motor can be obtained by integrating the sum of the slip angular velocity ω and the rotor angular velocity ω2 ((3) *
(4) electric model), and that φ can also be obtained by other means, such as a voltage model that calculates using the motor voltage and current, or a direct detection method. The aim is to indirectly obtain the rotor speed by matching the magnetic flux phases obtained by these two means.

Now, let us consider a case where the estimated speed value ω2 is used instead of the actual speed value ω2 to perform all magnetic flux calculations. If both of them match, the following equation (5) holds true.

At this time, the phase of the magnetic flux within the motor matches that calculated within the control circuit. Now, ω2 is Δω than ω2
Suppose that it becomes larger by 2. In this case, the angular velocity of the magnetic flux in the motor and that calculated in the control circuit are a deviation Δω2
Therefore, a deviation Δφ occurs in the phase of the magnetic flux obtained by integrating them. An adjustment loop that takes into account this deviation is designated by reference numeral 200 in FIG. In order to detect the magnetic flux inside the motor as accurately as possible, a voltage model method is used here in which the magnetic flux is calculated from the motor's primary voltage, secondary current, and motor constants.

Reference numeral 22 in FIG. 5 is a voltage model type magnetic flux calculator (also simply referred to as a voltage model). Here, the zero voltage model, which briefly explains the voltage model, calculates the secondary interlinkage flux F2 using the following equation, and an example of its configuration is shown in FIG.

(6) It can be seen from equation (7) that the magnetic flux can be obtained by correctly detecting the motor-large voltage and -order current vectors and accurately measuring the motor constants. In addition, the secondary interlinkage flux F2 is fixed on the α axis.

Component F on the β axis (coordinate system taken on the stator winding);
,,l), the following equation (8) is obtained.

When these are input into a known VA (vector analyzer) 23, the phase φ9 of the magnetic flux is determined as the following two-phase quantity.

These values are input to the vector rotator 24, and the deviation from the phase φ□ of the magnetic flux calculated using the current model is calculated using equation (11). Note that the phase φ1 of the magnetic flux is also (1
0) Expressed in the form of components as shown in the formula.

-sinψvcO3ψI CO3ψ, 5Lllψ Figure 8 shows an example of a vector diagram when a deviation occurs. Δ
In this case, the polarity of φ is positive in the forward rotation direction (counterclockwise direction) based on φI. Therefore, in the case of FIG. 8, Δφ is 0. The deviation Δφ obtained by equation (11) is determined by the controller 25.
The deviation is integrated here and the angular velocity ω7 is output. In the case of FIG. 8, regulator 25 integrates in the negative direction. This ω7 is added to ω at an addition point 103, and the output ω is integrated by an integrator 102, and the phase φ1 of the magnetic flux is determined by calculating equation (3). That is, adjustment loop 2
00, the regulator 25 operates so that this deviation Δφ becomes zero, and when the phases of the magnetic flux match, in principle, the rotational angular velocity ω1 of the magnetic flux in the electric motor and in the control device becomes (12)
Matches like expression.

ω1-ω9+ω5=ω2+ω, ・・・・・・
(12) Here, when Δφ is zero, torque? [, Since the command value and the actual value of the magnetizing current match, (4)
As is clear from the equation, the slip angular velocity command value ω3 matches the slip angular velocity command value ω8.

From the above, the output ω7 of the ml unit 25 becomes the estimated value ω2 of the rotational angular velocity.

[Problem that the invention seeks to solve]

In the example shown in FIG. 5, the voltage model shown by equations (6) and (7) is used for magnetic flux calculation. In the case of magnetic flux calculation using a voltage model, calculation accuracy is relatively good in a sufficiently high motor voltage range, but the accuracy deteriorates as the voltage decreases. The main cause of this is the voltage drop across the primary resistance (R1) included in the calculation equation. -The resistive voltage drop is typically a few percent of the rated motor voltage;
It is a constant value regardless of speed. Further, the induced voltage, which occupies a large proportion of the motor voltage, is proportional to the speed when the magnetic flux is constant. Therefore, as the speed decreases, the induced voltage becomes smaller, and on the other hand, since the -th order resistance voltage drop is a constant value, the ratio of the -th order resistance voltage drop to the motor voltage increases. Resistance generally tends to increase as temperature rises, so it is difficult to determine its exact value. Therefore, the setting accuracy of this resistance directly affects the accuracy of this magnetic flux calculation. In addition, voltage drops due to cables from the I/P to the motor and on-voltage drops of power semiconductors such as transistors and GTO thyristors are also major causes of errors. As described above, when the conventional voltage model is used, there is a problem in that the accuracy of magnetic flux calculation at low speeds and thus the accuracy of speed estimation decreases, resulting in a significant decrease in speed control performance.

Therefore, an object of the present invention is to provide a variable speed control device for an induction motor that enables high-performance speed control even at low speeds without providing a means for directly detecting speed.

[Means for solving problems]

The component of the motor primary voltage vector orthogonal to the current primary vector (the first
component) #! a calculation means for calculating a component (second component) orthogonal to the primary current vector of the motor primary voltage vector detected from the motor; and adjusting means for adjusting the II! The output of the node means is used as an estimated value of the motor rotational angular velocity.

[Effect]

In order to control the speed of an induction machine without a speed detector, the phase φ of the magnetic flux of the motor can be obtained by integrating the sum of the slip angular velocity ω5 and the rotor angular velocity ω2 ((a) t (4)
By focusing on the fact that φ can also be obtained from the motor voltage, the rotor speed can be obtained indirectly by matching the magnetic flux phases obtained by both methods, and even at low speeds, high speeds can be obtained. Enables speed control without a high-performance speed detector.

〔Example〕

FIG. 1 is a block diagram showing an entire embodiment of the present invention.

The basic principle for calculating the speed is the same as in the applied device, but the difference is that the 1pIWJ loop 300 is provided, that is, the input of the regulator for calculating the speed is different from the applied device.

FIG. 1 will be explained.

First, from the motor-secondary voltage vector v1 which is the output of the three-phase to two-phase converter 21, using the angle I from the fixed axis α of the current vector 11 (see FIG. 7) using the vector rotator 3°, A component v1δ (voltage invalid component) of the voltage vector v1 orthogonal to the current vector is calculated and inputted to the speed estimation circuit 32.

Note that the vector rotator 30 performs calculations as shown in the following equation.

・・・・・・(13) Here, Vl-(v, v).

1α 1β The current estimating circuit 32 calculates a voltage reactive component command value v1δ from the magnetizing current, torque current command values iM, IT, magnetic flux command value F2, magnetic flux angular velocity ω1, and motor constant, and calculates the voltage reactive component command value v1δ from this and the vector rotator. The internal controller performs a predetermined HffJ calculation so that the deviation ΔV□δ from the actual reactive voltage value v1δ given from 30 becomes zero, and the estimated speed value
Output.

This speed estimating circuit 32 also outputs an angle β (see FIG. 6), which is multiplied by the previously described φ□ at an addition point 31 to calculate an angle e. Also, the estimated speed value ωV(ω
7) is the same as in Figure 5, the Toguchi score is 103.
and given to 13. The configuration of N5 of the speed estimation circuit 32 will be described later.

The second part is a vector diagram for explaining the relationship between the motor primary voltage and the primary voltage. As is clear from the figure, the voltage reactive component v1δ is orthogonal to the current first-order vector 11, and is an amount unrelated to the voltage drop across the negative-order resistance (R1). Therefore, if the speed is estimated using this value, the magnetic flux of the motor and the magnetic flux of the current model can be matched even at extremely low speeds, and as a result, the speed can be calculated with high accuracy.

By the way, when the estimated speed value and the actual value deviate, a deviation occurs between the magnetic flux axis (M-axis) calculated within the control device (current model) and the magnetic flux axis (M-axis) within the electric motor. The relationship between the voltage and current vector at this time is shown in FIGS. 3A and 3B.

Note that FIG. 3A shows the case when the electric motor is in the driving state, and FIG. 3B shows the case when the electric motor is in the braking state, and both figures show the case where the M axis is angularly advanced compared to M@.

First, we will explain from Figure @3A. Now, assuming that the magnitude of the current flowing through the motor is constant, the M-axis component IM of this current is IM>1y1 in the case of the same figure.

This indicates that the motor magnetic flux is larger, and the magnitude of the induced->induced electromotive force also increases as lEz1>lEzl. Here, the motor constant is 81. When the primary voltage is determined assuming that the value of L does not change, its → relationship is lvl l>Ivl i, and the relationship of the δ-axis component is Vyuδ>V□δ. In this way, when a shift occurs in the magnetic flux axis, the change affects the δ-axis component of the voltage vector. Therefore, by focusing on and adjusting the δ-axis component of the voltage vector, it becomes possible to eliminate the deviation of the magnetic flux axis. Note that in order to make the M-axis coincide with the M-axis in the figure, it is necessary to rotate the M-axis counterclockwise, which corresponds to increasing ω7 in FIG. 1.

! @ In the case of figure 3B, the M-axis is ahead of the M-axis as in the case of figure 3A, so it is necessary to rotate the M-axis counterclockwise. However, in this case, from the relationship in the same figure, Vi
δ(V+δ), and the polarity of the deviation ΔV□δ is different from that shown in Figure 3A, so the polarity of the deviation Δv1δ is reversed when braking.The braking and driving states can be determined using the γ axis of the voltage vector. This can be easily done using the component (Vl, ).That is, since this r-axis component is an effective component of the voltage vector, it is positive during driving and negative during braking.

In this way, when calculating speed, we use driving.

Although it is necessary to consider braking, in any case, by focusing on the voltage reactive component, it is possible to eliminate the misalignment of the two magnetic flux axes.

FIG. 4 is a block diagram showing a specific example of the speed estimation circuit. In the figure, 321 is a well-known system that performs polar coordinate transformation.
p converter, 322 is a v1δ calculator, 323 is an addition point, 3
24 is a polarity switching circuit, and 325 is a regulator.

That is, the k/p converter 321 changes the torque current command value I
Convert T and magnetizing current command IM into polar coordinate quantities. The relational expression is as follows.

β-taII (i7/IM)・”・
(14) From these quantities, the rotational angular velocity ω of the magnetic flux, the magnetic flux command value F2, and the motor constant, the calculator 322 calculates the voltage reactive component command value v1δ. This is the voltage reactive component v1
δ1X.

1. M v1δ−ωr La ' □” M+12((1”z
) X(-siaβ)-+-ω1F2 CO3β)...
... (16) This is because it is known to be given as follows.

Note that ω in equation (16) is ω, = ω, +ωβ, and ωβ
The quantity expressed as 1dβ/dt, L(F * M @ 1
2 indicates the motor constant, p is the differential operator (d/d t
) is shown.

Also, equation (16) is expressed as ωβ when the torque change rate is small.
Since it is considered to be -0, (-5inβ) +ω1F2 cosβ)...
(17), and when the magnetic flux change rate is small, (pF2
): O, so it is simplified as (18).

The voltage reactive component command value v1δ obtained in this way is input to the polarity switching circuit 324 after the deviation Δv1δ from its actual value v1δ given from the vector rotator 30 is taken at the power calculation point 323. voltage component v1r
Driven by,? blJ motion discrimination All discrimination polarity switching is performed. The output of the polarity switching circuit 324 is input to the regulator 325, and an estimated speed value ωV is obtained.

〔Effect of the invention〕

According to the present invention, high-performance control equivalent to the case with a speed detector can be realized without using a speed detector. In addition, without changing the basic configuration of vector control with speed control, high-performance control without a speed detector is possible by simply adding a control loop. Therefore, the present invention is suitable for applications where a speed detector cannot be attached, or when speed signals are used. It is particularly effective in applications where it is not possible to transmit light to the glass.

[Brief explanation of the drawing]

Figure 1 is a configuration diagram showing an embodiment of the present invention, Figure 2 is a vector diagram showing the relationship between the motor primary current and primary voltage, and Figure 3A is the magnetic flux axis in the control circuit and the magnetic flux in the motor when the motor is driven. Figure 3B is a vector diagram to explain the deviation between the magnetic flux axis in the control circuit and the magnetic flux axis in the motor during motor braking. FIG. 5 is a block diagram showing a specific example, FIG. 5 is a configuration diagram showing an ejected device, FIG. 6 is a vector diagram showing the current of an induction motor, FIG. 7 is a circuit diagram showing a specific example of a current model, and FIG. 8 is a block diagram showing a specific example. is a vector diagram for explaining the phase deviation of each magnetic flux obtained by a current model and a voltage model. Description of symbols 1... PWM inverter, 2... Induction motor, 3... Differential circuit, 4... Magnetic flux command calculator, 5...・Speed regulator (ASR), 6
...Magnetizing current regulator (ACR), 7...
・Torque current iam device, 8... Coordinate conversion circuit,
9...Pulse generation circuit, 10...Current model type magnetic flux calculator, 11, 24.30...Vector rotator (VD), 12.21...・・3 phase-
2-phase converter, 13, 14. 15, 31, 103, 323
... Addition point, 20 ... Voltage transformer, 2
2b, 22c, 22d...Coefficient, vessel, 23...
...Vector analyzer, 25, 325---
--'IIWJ unit, 22a*102...Integrator, 32...Speed estimation circuit, 100...
・Speed setter, 101...Slip angular velocity calculator, 200.300...Adjustment loop, 321...
. . . k/p converter, 322 . . . vIJ+ arithmetic unit, 324 . . . polarity switching circuit. Agent Patent attorney Akio Namiki Agent Patent attorney Kiyoshi Matsuzaki 2 Figure C'M4 Figure 3A Figure 3B γ simple

Claims (1)

[Claims] The primary current of an induction motor, which is supplied via a power converter capable of controlling the magnitude, frequency and phase of the output voltage, is divided into a component parallel to the magnetic flux of the motor (magnetizing current) and a component parallel to the magnetic flux of the motor (magnetizing current). In a variable speed control device for an induction motor that controls at least the motor torque by separating the component (torque current) orthogonal to component of the motor primary voltage vector (first
a first calculation means for calculating a component (component); and a second calculation means for calculating a component (second component) orthogonal to the primary current vector of the motor primary voltage vector detected from the motor.
and an adjusting means for adjusting the deviation between the first component and the second component to zero, and the output of the adjusting means is used as an estimated value of the motor rotation angular velocity. Variable speed control device for induction motor.