JPH06101958B2 - High-speed torque controller for induction motor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial application] The present invention relates to instantaneous magnetic flux and torque control of an induction motor by an inverter used for high-speed digital control, and aims to increase the capacity of the system. It is intended to use GTO with a limited switching frequency, and it can be applied to AC servomotors that do not necessarily aim for large capacity but require high-speed torque control.

[Conventional technology]

At present, vector control is widely used when high-speed torque control is required for induction motors. However, this control method is highly dependent on the internal constants of the motor, and requires complicated position adjustments such as a position sensor, and has the drawback that changes in the internal constants of the motor greatly affect the control characteristics. There is. Further, since the current control is the basis, the instantaneous magnetic flux and the instantaneous torque are not necessarily optimized, and there are disadvantages such as wasteful switching.

As a new control method replacing this control, the present inventors have previously developed a new magnetic flux calculation type control method. Regarding this, SPC-85-45 “High-speed and high-efficiency control method for induction machines based on new logic” (Noguchi, Takahashi) published in the Institute of Electrical Engineers of Japan Semiconductor Power Conversion Study Material published in August 1985 and Showa era 61
Vol. 106, No. 1 of the Transactions of the Institute of Electrical Engineers of Japan published in January 6
1-B2 "New high-speed torque control method for induction motors based on instantaneous slip frequency control" (Takahashi, Noguchi), etc., but it will be briefly described below.

FIG. 8 is a block diagram of an example of a magnetic flux calculation type control method.
The inverter part is shown in a simplified manner. That is, the inverter 3 is composed of six arms each having a switching element such as a transistor and a diode connected in antiparallel, and as shown in the figure, three changeover switches _Su ,
It can be expressed as S _v and S _w .

The three-phase voltage type inverter 3 has a positive bus from the DC voltage source 1.
Is fed via the 1a and negative bus 1b, the selector switches S _u of the inverter 3 by the control circuit 7, S _v, S _w is positive, negative bus
The AC power converted by falling to the 1a, 1b side,
Power is supplied to each phase terminal u, v, w of the three-phase induction motor 6 via the current detectors 5 _u , 5 _v , 5 _w . The voltage of the DC voltage source 1 is detected by a voltage detector 2 inserted between the positive and negative buses.

In this control method, the electromagnetic quantity is treated as an instantaneous vector represented by orthogonal dq2 axes. That is, the primary voltage Primary current Secondary current The primary magnetic flux And It is represented by. Here, j represents a vector product.

From the three-phase voltages v _u , v _v , v _w and the three-phase primary currents i _u , i _v , i _w Can be calculated by the following equations.

The voltage equation of the induction motor 6 is Where R ₁ ; primary winding resistance L ₁₁ ; primary inductance R ₂ ; secondary winding resistance L ₂₂ ; tertiary inductance M; mutual inductance, where _m is the rotational angular velocity and p is the differential operator.

On the other hand, as the definition of magnetic flux, Expand the first line of the expression Substituting an expression into this and rearranging If you integrate both sides That is, Is calculated by the integral calculation of the formula.

On the other hand, the instantaneous torque T is It is calculated by the formula as the vector product of.

Right side of expression Is generally very small compared to You can think of it as extending in the direction of.

Is the output voltage of the inverter 3.

The output voltage vectors of the inverter 3 in the eight switch states are expressed by the formulas when the changeover switches constituting the inverter 3 fall to the positive bus line 1a side and 0 when they fall to the negative bus line 1b side, respectively. When calculated, it is as shown in the following voltage vector table. However, the actual values of v _1d and v _1q are the values in the table. And the voltage E of the DC voltage source 1 detected by the voltage detector 2 are multiplied.

When the switch state number is k, the inverter 3 in each switch state Is the same as the d-axis as shown in the voltage vector diagram of FIG. And then proceed 60 degrees clockwise And two zero vectors It will be 8 kinds of.

Blocks 701 and 703b included in the control circuit 7 of FIG.
Is a block which calculates the change-over switch S _u, S _v, S _w of the state and the voltage detector by 2 voltage E Tokara expression of the DC voltage source 1 detected by the primary voltage ▲ ▼ a.

The block 702 is a three-phase current i _u , i _v , i _w detected by the current detectors 5 _u , 5 _v , 5 _w Is a block for calculating.

this Is multiplied by the primary winding resistance R ₁ in block 703a and in block 704 From the primary winding resistance R ₁ Subtract the product of.

A block 705 is a block for integrating the magnetic flux according to the formula, The d and q biaxial components φ _1d and φ _{1q of} are obtained.

In block 710 Depending on which region the clockwise rotation angle θ with respect to the d-axis of, belongs to the boundaries, which are divided by 60 ° of 30 °, 90 °, 150 °, 210 °, 270 °, 330 °. Control flag f
θ is generated as follows.

−30 ° ≦ θ <30 °; fθ = I 30 ° ≦ θ <90 °; fθ = II 90 ° ≦ θ <150 °; fθ = III 150 ° ≦ θ <210 °; fθ = IV 210 ° ≦ θ < 270 °; fθ = V 270 ° ≦ θ <330 °; fθ = VI block 706 Is a block for calculating by the following equation.

Externally provided at block 708 Is subtracted to calculate the deviation of the magnetic flux.

A block 711 is a hysteresis comparator. When the deviation of the magnetic flux sent from the block 708 is positive and exceeds a predetermined value, that is, when it is necessary to increase the magnetic flux, the control flag fφ = + 1 is set, and the deviation of the magnetic flux is negative and the predetermined value. Is exceeded, that is, when it is necessary to reduce the magnetic flux, the control flag fφ = −1.

A block 707 is a block for calculating the instantaneous torque T by calculating the vector product of both outputs of the blocks 702 and 705, and subtracting the instantaneous torque T from the torque command value T ^* given from the outside in block 709 to obtain the torque deviation. To calculate.

Block 712 is a ternary hysteresis comparator,
When the torque deviation is within a predetermined error range, a zero vector mode control flag fτ = 0 is generated, and when the torque deviation is positive and exceeds a predetermined value, that is, when acceleration torque is required, the control flag fτ = + 1 is set, and the torque When the deviation is negative and exceeds a predetermined value, that is, when deceleration torque is required, the control flag fτ = -1.

A block 713 is a block for determining the most suitable inverter output voltage for each combination of the three control flags fθ, fφ, fτ output from the blocks 710, 711, 712, and according to the control flags fθ, fφ, fτ from the switching table shown below. Switch state number k shown in the previous voltage vector table
Then, a switching signal is sent to the inverter 3 to instantaneously control the magnetic flux and the torque.

As described in detail above, according to such a magnetic flux calculation type control method, each moment is calculated while using almost no internal constant of the induction motor. Is kept almost constant, and the Lissajous figure of the magnetic flux vector draws a circle, and high-speed torque control can be performed.

[Problems to be solved by the invention]

As described above, the high-speed torque control method is an extremely high-performance magnetic flux calculation type control method. However, even in this conventional control method, inconvenience may occur in the case of low speed and light load, and large transient fluctuations may occur. However, there is a drawback that special treatment is required to deal with the torque.

That is, there are only six inverter output voltage vectors available, excluding the zero vector, and the degree of freedom in control is limited.

Therefore, since a voltage vector having a magnetizing component and a small torque component is not always obtained, when the speed is low and the load is low, the zero vector is often selected and the magnetizing cannot be performed. Therefore, the amount of magnetic flux decreases in the low speed range, The Lissajous figure is distorted into a hexagon, which causes noise and vibration.

In this conventional control method, when an attempt is made to apply a high voltage to obtain a high acceleration, the voltage of the DC voltage source 1 is generally constant, so the value becomes lower than the maximum voltage of the inverter determined by this input voltage. The rated voltage of the induction motor must be set and the inductance of the induction motor must be low.

However, in this case, the pulsation of the motor current constantly increases, the pulsation of the torque also increases, and the iron loss and the magnetic noise both increase.

On the contrary, if the current pulsation is constantly kept low to reduce the torque pulsation, iron loss, and magnetic noise, the rated voltage of the induction motor is set to a value almost equal to the inverter input voltage, and the impedance of the induction motor is also set to a large value. Need to be designed so that

In this case, a large torque cannot be transiently generated.

Therefore, there has been a demand for a high-speed torque control system for an induction motor, which can transiently generate a large torque and can constantly realize low torque pulsation, low loss, and low noise.

[Means for solving problems]

The present invention relates to a control system using an inverter of a three-phase induction motor, in which terminals a ¹ a ² , b ₁ b ₂ and c ₁ are provided at both ends, respectively.
A three-phase induction motor having three-phase symmetrical windings electrically isolated from each other having c ₂ is used.
Of the terminal groups a _1, b _1, c ₁ while connected to the first 3-phase voltage source inverter, a second three-phase reactor terminal group a _2, b _2, c ₂ for the zero-phase current suppression To the second three-phase voltage source inverter via the first and second voltage sources so that the instantaneous magnetic flux and the instantaneous torque of the three-phase induction motor follow a command value given within a predetermined error range. Voltage control means for selecting the inverter output voltage, and zero for selecting the conduction mode of the first and second voltage source inverters that reduces the current value of the zero-phase current from the plurality of conduction modes of the voltage control means output. It is characterized in that it is provided with a phase current suppressing means.

The three-phase reactor for suppressing the zero-phase current is composed of three-phase windings wound around a common iron core and having magnetic coupling in the same direction, and operates as a reactor only for current components in the same direction. Use the one that

[Action]

FIG. 1 is a main circuit connection diagram in a high-speed torque control device for an induction motor according to the present invention. The three-phase induction motor 10 comprises electrically isolated three-phase symmetrical windings having terminals a ₁ a ₂ , b ₁ b ₂ and c ₁ c ₂ at both ends.

The greatest feature of the system according to the present invention is to use an induction motor having a three-phase symmetrical primary winding electrically insulated in this way. In this figure, a star-shaped connection with an open neutral point is used. However, any complex winding may be used as long as it is a three-phase symmetrical winding.

The first terminal group a ₁ , b ₁ , c ₁ of the three-phase symmetrical winding is connected to the first three-phase voltage source inverter 11, and the second terminal group a ₂ ,
b ₂ and c ₂ are connected to the second three-phase voltage source inverter 12 via the three-phase reactor 13 for suppressing the zero-phase current.

In such a system, the zero-phase current i _o = i _a + i _b that does not affect the characteristics of the motor but increases the line current.
+ I _c exists. This is also suppressed by the primary-side leakage inductance of the induction motor, but its value is as small as several mH, so the three-phase reactor 13 for suppressing the zero-phase current is inserted.

The first and second three-phase voltage source inverters 11 and 12 are
Similar to the description of the conventional example of FIG. 8, three changeover switches S _1a , S _1b , S _1c and S _2a , S _2b , S _2c are shown. Both three-phase voltage source inverters are DC voltage source 1
From the positive bus 1a and the negative bus 1b.

Here, as explained in the conventional example, each changeover switch S _1a , S
Positive side busbar 1a for _1b , S _1c and S _2a , S _2b , S _2c respectively
When falling to the side, 1 when falling to the side of the negative bus 1b is represented by 0, and the switch status numbers of the double voltage source inverter are respectively indicated.
In terms of k ₁ and k _2, there are 8 different switch states, as shown in the following switch state table.

Since each of the first and second three-phase voltage source inverters 11 and 12 has eight switching states, it can be seen that the combination of these has 64 switching states in this system.

The voltage v _a , v _b , v _c applied to each phase winding having the terminals a ₁ a ₂ , b ₁ b ₂ , c ₁ c ₂ of the three-phase induction motor 10 is a changeover switch connected to each winding. It changes depending on the switch states of S _1a and S _2a , S _1b and S _2b , and S _1c and S _2c .

For example, considering the phase a, the voltage of the DC voltage source 1 is E
, S _1a and S _2a are both positive bus 1a side or negative bus 1b
If it falls to the side, v _a = 0, S _1a falls to the side of the positive bus 1a S
_2a is when the fallen to negative bus 1b side v _a = E, if S _1a is S _2a lying on negative bus 1b side is lying on positive bus 1a side v _a
= -E.

For each combination of the switch state numbers k ₁ and k ₂ of the first and second three-phase voltage source inverters 11 and 12, the voltage v _a , v _b , v _c applied to each phase winding, and d− at that time There are 19 types of composite voltage vectors that have undergone q2-axis conversion, and the distinction is shown as V _{0 to} V ₁₈ by the lower right suffix of V.
It can be seen that a plurality of switch states exist for the same combined voltage vector, and similarly, the combined voltage vectors V _{0 to} V ₆ have different phase voltages even if they are the same combined voltage vector.

Resultant voltage vector is = is represented by V _d + jV _q, can be obtained each phase winding voltage v _a, v _b, from v _c by the following equation.

The magnitude of the composite voltage vector is the same as that of the conventional three-phase induction motor drive system using the inverter described in FIG. Of the And 1.5 times.

V _{1 to} V ₆ are conventional , V _{13 to} V ₁₈ are ▲ √ ▼ times, and V _{7 to} V ₁₂ are 1.5 times. The small combined voltage vectors V _{1 to} V ₆ are mainly used for steady state control, and the other large combined voltage vectors are used for transient state control.

The torque and the magnetic flux are controlled by selecting the combined voltage vector as in the conventional system, but the zero-phase current is suppressed by switching the phase voltage state of the combined voltage vector or changing the combined voltage vector.

It may be necessary to change the composite voltage vector to V ₁₃
A case where ~V ₁₆ is selected, this is despite the synthesis voltage vector V ₁₃ ~V ₁₈ is zero-phase voltage is represented by _{v 0 v 0 = v a +} v b + v c is not zero, the phase This is because there is only one voltage state.

The characteristic equation on the primary side of the three-phase induction motor 10 is expressed as = R ₁ + P …………………………. Where R ₁ is the primary winding resistance, is the primary magnetic flux,
Is the primary current vector and is expressed as = I _d + jI _q , and from each phase line current i _a , i _b , i _c Can be calculated by

Primary magnetic flux = Φ _d + jΦ _q is also expressed by the formula = ∫ (-R ₁ ) dt …………………………, and if the voltage drop of the winding is neglected, the locus of the primary magnetic flux vector Is in the direction of the composite voltage vector,
Moves at a speed proportional to the size of. Therefore, if the combined voltage vector is properly selected, the absolute value || of the primary magnetic flux can be controlled to be kept constant.

FIG. 3 is a diagram showing the locus of the primary magnetic flux vector, and shows the relationship between the locus of the primary magnetic flux vector and the combined voltage vector when the primary magnetic flux vector is rotated clockwise. When | Φ ^* | is the magnetic flux command value and | ΔΦ | is the allowable error, the combined voltage vector is selected so that | Φ ^* |-| ΔΦ | ≦ || ≦ | Φ ^* | + | Δφ | is satisfied. To be done.

Similarly to the equation, the torque T can be calculated by T = Φ _d × I _q −Φ _q × I _d ………………, and can be obtained only by the internal constant on the primary side.

Fig. 4 is a graph showing the step response of torque with respect to the slip angular frequency, which was published in the above-mentioned document "High-speed and high-efficiency control method for induction machines based on the new theory" and shows the primary magnetic flux for the rotor. Slip angular frequency which is relative velocity
_It shows the step response of torque when _s is stepwise changed.

According to this figure, the steady-state torque has a maximum value at _s = 2π · 15, but it is understood that the rate of increase in torque during the transition is almost proportional to the magnitude of the slip angular frequency _s .

From such torque characteristics, assuming that the change in the rotor angular velocity _m is very slow with respect to the change in the torque T, the torque can be controlled by changing the rotational angular velocity _φ of the primary magnetic flux vector.

At this time, if the slip angular frequency _s = _φ − _m increases, the torque T increases, and if _s decreases, the torque T also decreases. Thus, the angular velocity of rotation of the primary magnetic flux vector
Limit cycle control of the torque T is possible by changing _φ .

Next, a method for selecting a combined voltage vector for simultaneously controlling torque and magnetic flux will be described. FIG. 5 is a diagram showing the relationship between the combined voltage vector and the primary magnetic flux vector at a certain time.

From the formula, the combined voltage vector is almost 90 ° from the primary magnetic flux.
The primary magnetic flux vector can be rotated in the direction of the arrow θ by selecting v ₀ , v ₂ , v ₃ , v ₈ , and v ₁₅ . Among these, the one that decreases the primary magnetic flux vector is V ₃ to V ₁₅ , and the one that increases it is V ₂ or V ₈ .

Further, the rotational angular velocity φ of the primary magnetic flux vector is minimum in V ₀ , intermediate in V ₂ and V ₃ , and maximum in V ₈ and V ₁₅ .

Therefore, when the average value of the rotational angular velocity φ of the primary magnetic flux vector is low, when the torque T is decreased, V ₀ , when the torque T is increased and the primary magnetic flux is increased,
Select V ₂ and also V ₃ to reduce the primary magnetic flux vector.

When the average value of the rotational angular velocity φ of the primary magnetic flux vector is high, select V ₂ to increase the primary magnetic flux vector and reduce the torque T, and select V ₈ to increase the torque T. However, when the primary magnetic flux vector is reduced and the torque T is reduced, V ₃ and torque T
Select V ₁₅ to increase.

Although the combined voltage vector to be selected changes depending on the current position of the primary magnetic flux vector, the above-mentioned idea may be used. The region to which the primary magnetic flux vector belongs is divided by the rotation angle θ in the clockwise direction with respect to the d axis of the primary magnetic flux vector. In the conventional example shown in FIG. 8, there are six areas, but in the present invention, the directions of the selectable combined voltage vector are increased. Therefore, as the area signal θ _n , ₁₂ areas of θ ₁ to θ ₁₂ are set as the rotation angles below. It emits in the range of θ.

0 ° ≦ θ <30 °; θ n = θ 1 30 ° ≦ θ <60 °; θ n = θ 2 60 ° ≦ θ <90 °; θ n = θ 3 90 ° ≦ θ <120 °; θ n = θ ₄ 120 ° ≦ θ <150 °; θ _n = θ ₅ 150 ° ≦ θ <180 °; θ _n = θ ₆ 180 ° ≦ θ <210 °; θ _n = θ ₇ 210 ° ≦ θ <240 °; θ _n = θ ₈ 240 ° ≤ θ <270 °; θ _n = θ ₉ 270 ° ≤ θ <300 °; θ _n = θ ₁₀ 300 ° ≤ θ <330 °; θ _n = θ ₁₁ 330 ° ≤ θ <360 ° Θ _n = θ ₁₂ Absolute value of primary magnetic flux || and torque T are compared with magnetic flux command value | Φ ^* | and torque command value T ^* by hysteresis comparator, and when magnetic flux increase / decrease signal φ is increased, 1, When demagnetizing, 0 is output and torque increase / decrease signal τ ₁
As a result, 1 is output when the torque is increased, 0 is output when the torque is decreased, and the rotational angular velocity φ of the primary magnetic flux vector is output.
By determining the direction and the magnitude of the above, it is possible to select a desired combined voltage vector by the combined voltage vector selection table shown below according to the area signal θ _n .

As described above, in such a system, the zero-phase current i ₀ = i _a + i _b + i _c may flow, which increases the loss. Therefore, the zero-phase current i ₀ is suppressed. There is a need.

The following relational expression holds between the zero-phase current i _{0 and} the zero-phase voltage v ₀ .

Where R ₀ ; winding resistance of 3-phase reactor 13 R ₁ ; primary winding resistance of induction motor 10 L ₀ ; leakage inductance of 3-phase reactor 13 L ₁₁ ; primary inductance of induction motor 10 M ₀ ; 3-phase mutual inductance of the reactor 13 Therefore, the zero-phase current i ₀ can be controlled by the zero-phase voltage _{v 0 = v a + v b} + v c.

That is, it is only necessary to select, from the combined voltage vectors selected by the above-mentioned combined voltage vector selection table, the one in the conduction mode having the phase voltage that produces the zero phase voltage component in the opposite direction to the zero phase current i ₀ .

Actually, the zero-phase current i ₀ is detected and compared with a predetermined maximum zero-phase current Δi ₀ by a three-value hysteresis comparator, and when the maximum zero-phase current Δi ₀ or less is set as the zero-phase current control signal I ₀ , 0 , -1 is generated when the value exceeds the positive value, and +1 is generated when the value exceeds the negative value, and the zero-phase current control phase voltage table shown below is used by the zero-phase current control signal I ₀ and the region signal θ _n . Select the optimal energization mode.

Zero-phase current control signal I _{0 for} combined voltage vector V _{1 to} V ₆
To switch between the two phase voltage states. Since the zero-phase voltage v _{0 of the} combined voltage vectors V ₀ and V _{7 to} V ₁₂ is zero, the zero-phase current i
₀ does not change. Since the combined voltage vectors V _{13 to} V ₁₈ have only one phase voltage state, when the zero-phase current control signal I ₀ commands switching, the primary or secondary magnetic flux vector area signal θ _n at that time is rare or even. Depending on the combined voltage vector V ₇ ~ V ₁₂
Switch to.

Since this determines the phase voltage that suppresses the zero-phase current i ₀ , it must be converted into a switching signal. A changeover switch connected to any one-phase winding of a three-phase induction motor
When S _1x and S _2x are set, the switch states are uniquely determined as (1,0) and (0,1) when the phase voltages are E and −E. However, since the switch state when the phase voltage is 0 may be either (0,0) or (1,1), these two changeover switches S _1x and S _2x are switched so as to be equalized.

〔Example〕

FIG. 7 is a block diagram of an embodiment of the high-speed torque control device for an induction motor according to the present invention.
The phase voltage source inverters 11 and 12, the zero-phase current suppressing three-phase reactor 13 and the three-phase induction motor 10 are the same as those shown in FIG. 1, and are shown as a single wire connection for simplification. is there.

The DC voltage E is supplied from the DC voltage source 1 to the first and second three-phase voltage source inverters 11 and 12, and the output of the first three-phase voltage source inverter 11 is the three-phase induction motor 10 with the neutral point opened. Connected to the first terminal group a ₁ , b ₁ , c ₁ which are the normal side terminals of the second three-phase voltage source inverter 12 output is the open neutral point side terminal of the three-phase induction motor 10. Second terminal group
It is connected to a ₂ , b ₂ , and c ₂ via a three-phase reactor 13.

The winding voltage v _a , v _b , v _c and the line currents i _a , i _b , i _c are fed back from the winding voltage detecting means 21 and the line current detecting means 22 for each phase, and each of them is converted into three-phase to two-phase. Means 23 and
At 24, the d- _q biaxial components are converted into V _d , V _q and I _d , I _q .

The voltages V _d and V _q are respectively the primary winding resistance simulating means 25 and 26.
To reduce the primary voltage drops R ₁ I _a and R ₁ I _q by subtracting means 27 and
After being subtracted by 28, the d-axis component Φ _d and the q-axis component Φ of the primary magnetic flux expressed by the equations by the integrating means 29 and 30.
_q is calculated.

Both axis components Φ _d and Φ _q of the primary magnetic flux are the magnetic flux region determination means.
_One of θ ₁ to θ ₁₂ is generated as the area signal θ _n and is also sent to the magnetic flux absolute value calculating means 32.
Absolute value of the next magnetic flux || Is calculated by

Subtracting means 33 from externally applied magnetic flux command value | Φ ^* |
At, the absolute value || of the primary magnetic flux is subtracted, and the error of the magnetic flux is sent to the hysteresis comparator 34 to generate the magnetic flux increase / decrease signal φ. The magnetic flux increase / decrease signal φ is either 1 or 0. When 1 is set, the magnetic flux is increased and when 0 is set, the demagnetization is instructed.

The biaxial components Φ _d and Φ _q of the primary magnetic flux are further the biaxial components I _q of the line current output from the 3-phase to 2-phase conversion means 24, respectively.
And I _d are multiplied by multiplying means 35 and 36, and subtracting means 37
The torque T according to the formula is calculated by subtracting with.

The torque T is subtracted from the torque command value T ^* given from the outside by the subtracting means 38, and the error of the calculated torque is the hysteresis comparator 39 and the comparators 40 and 41.
Sent to.

The hysteresis comparator 39 sets the error limit ΔT ₁ to about 2% of the torque command value, and the torque error is the error limit ΔT ₁
Is exceeded on the positive side, 1 is generated as the torque increase / decrease signal τ ₁ to instruct to increase the magnetism, and when it is exceeded on the negative side, 0 is generated as the torque increase / decrease signal τ ₁ to instruct demagnetization.

Comparators 40 and 41 each have 5% of the torque command value
There is an error limit ΔT _{2 of} about 10% and an error limit ΔT _{3 of} about 10% of the torque command value. In both cases, the torque difference signals τ ₂ and τ ₃ are set to 1 when the input exceeds the error limit, and 0 is set within the error limit. Send to the up / down counter 42.

The up / down counter 42 receives the torque increase / decrease signal τ ₁ and the torque difference signals τ ₂ and τ ₃ as inputs, and outputs 1 to 4 as the error signal d.
When the torque difference signal τ ₂ changes from 0 to 1, when the torque increase / decrease signal τ ₁ is 1, the count is incremented by 1 with 4 being the upper limit, and the torque increase / decrease signal τ ₁ is 0.
In this case, 1 is counted down by 1 with 1 being the lower limit. When the torque difference signal τ ₃ changes from 0 to 1, when the torque increase / decrease signal τ ₁ is 1, the content is set to 4 and when the torque increase / decrease signal τ ₁ is 0, regardless of the content. The content is 1.

The content of the up / down counter 42 becomes an error signal d. There are four types of error signals d 1 to 4, and these correspond to the rotation speed φ of the primary magnetic flux vector described in the above composite voltage vector selection table. When φ >> 0, d = 1 φ> 0, d = 2 φ <0, d = 3 φ << 0, d = 4.

On the other hand, from the line current detecting means 22, the line currents i _a , i _b , i _c of each phase
Is also sent to the summing means 43, and the zero-phase current i ₀ is calculated and sent to the ternary hysteresis comparator 44.

The ternary hysteresis comparator 44 sets the operation limit Δi ₀ and generates the zero-phase current suppression signal I ₀ as an output. The zero-phase current suppression signal I ₀ is −1 when the zero-phase current i ₀ exceeds the operation limit Δi ₀ on the positive side, +1 when the zero-phase current i ₀ exceeds the operation limit Δi 0, and 0 when it is less than or equal to the operation limit Δi ₀ .

The storage means 45 is mainly composed of a ROM storing the above-mentioned combined voltage vector selection table and zero-phase current suppressing phase voltage table, and includes a torque increase / decrease signal τ ₁ , an error signal d, and a magnetic flux increase / decrease signal φ. , Region signal θ _n and zero-phase current suppression signal I ₀
, The optimum energization mode is selected from the combined voltage vector selected from the combined voltage vector selection table, and is output in the form of the switching signal of each switching element of the first and second inverters 11 and 12.

The switching equalizing means 46, which has received the output of the storage means 45, operates so as to equalize the switch states of both switching elements whose phase voltage becomes 0 as described above. Storage means 45
By sending the output of the output change detection means 47 which generates an output when the output of the change to the latch means 48, by returning the previous output of the switching equalization means 46 held by the latch means 48 to the switching equalization means 48, Switching equalization is performed.

〔The invention's effect〕

As described in detail above, the 19 kinds of combined voltage vectors generated by the first and second two voltage source inverters are selectively used to suppress the zero-phase current and equalize the switching, while transiently increasing the speed. Has a torque response of
In addition, it is possible to construct a drive system for an induction motor that has a low torque ripple and low noise in a steady state.

The high-speed torque control device for an induction motor according to the present invention is GT
Using a switching element such as O that does not increase the switching frequency too much enables a system with a large capacity and a high-speed response, and at the same time, if a high-speed switching element is used, the transient response and the It is also possible to use a servo system that also has steady characteristics.

[Brief description of drawings]

FIG. 1 is a main circuit connection diagram in a high-speed torque control device for an induction motor according to the present invention, FIG. 2 is a combined voltage vector diagram, FIG. 3 is a diagram showing a locus of a primary magnetic flux vector, and FIG. 4 is a slip angle. FIG. 5 is a graph showing the step response of torque with respect to frequency, FIG. 5 is a diagram showing the relationship between the combined voltage vector and the primary magnetic flux vector, FIG. 6 is a conventional voltage vector diagram, and FIG.
FIG. 8 is a block diagram of an embodiment of a high-speed torque control device for an induction motor according to the present invention, and FIG. 8 is a block diagram of an example of a conventional magnetic flux calculation type control method. 1 ... DC voltage source, 2 ... voltage detector, 3 ... inverter, 6,10 ... three-phase induction motor, 7 ... control circuit, 11 ...
First three-phase voltage source inverter, 12 ... second three-phase voltage source inverter, 13 ... three-phase reactor.

Claims (5)

[Claims] 1. A torque control system using an inverter for a three-phase induction motor, comprising a first group of terminals for a three-phase induction motor having electrically isolated three-phase symmetrical windings having terminals a1a2 b1b2 c1c2 at both ends. Connect a1 b1 c1 to the first three-phase voltage source inverter, and also connect the second terminal group a2 b2 c2 to the second three-phase voltage source inverter via the three-phase reactor for zero-phase current suppression. , So that the instantaneous magnetic flux and the instantaneous torque of the three-phase induction motor follow the given command value within a predetermined error range.
A means for determining the primary magnetic flux beltle from the primary voltage and the primary current of the phase induction motor, and determining in which circular arc area that the primary magnetic flux vector is divided into 12 equal parts, the area signal θ _n is calculated. A magnetic flux region determining means for outputting; a means for outputting a magnetic flux increasing / decreasing signal φ for instructing to increase or decrease the magnetization in accordance with a deviation between the absolute value of the primary magnetic flux vector and its command value; Torque T from primary voltage and primary current
And a torque increase / decrease signal .tau.1 for instructing torque increase when the torque error exceeds the error limit on the positive side and for decreasing the torque when the torque error exceeds the error limit on the positive side. Means for generating a torque difference signal τ2τ3 according to the torque error, and means for outputting an error signal d evaluated according to a combination of the torque increase / decrease signal τ1 and the torque difference signal τ2τ3. A means for calculating the zero-phase current i0 by summing the line currents of the respective phases, and a zero-phase current suppression signal I0 when the zero-phase current i0 exceeds or does not exceed a predetermined value on the positive and negative sides. A three-value hysteresis comparator for outputting the torque increase / decrease signal τ1, the error signal d, and the magnetic flux increase / decrease signal φ.
From the switching table stored in advance according to the combination of the region signal θ _n and the zero-phase current suppression signal I0.
And a storage means for outputting a switching signal of the second three-phase voltage source inverter, the high-speed torque control device for an induction motor. 2. The three-phase reactor is composed of three-phase windings wound around a common iron core and having magnetic coupling in the same direction, and operates as a reactor only for current components in the same direction. A high-speed torque control device for an induction motor according to the range (1). 3. The output signals of the storage means are first and second.
A high-speed torque control device for an induction motor according to claim (1), which is output as an instantaneous voltage vector format generated by the three-phase voltage source inverter. 4. The high-speed torque of an induction motor according to claim 1, wherein the three-value hysteresis comparator selects a conduction mode for reducing a zero-phase current represented by a vector sum of instantaneous currents of respective phases. Control device. 5. A high-speed torque control device for an induction motor according to claim 1, wherein the switch state output from said storage means is stored and a switch state different from the previous one is selected for next switching.