JPS63228991A - Torque magnetic-flux controlling system for induction motor - Google Patents

Abstract

PURPOSE:To reduce torque ripple at the time of a low speed and a light load, by integrating primary magnetic flux with motor terminal voltage, and by controlling magnetic flux and torque independently at the same time. CONSTITUTION:In the block 601' of a controlling circuit 6', voltage detectors 7u, 7v, 7w detect phase voltage va, vb. Taking into consideration phase currents ia, ib detected by current detectors 4u, 4v, 4w and the voltage drop across the primary winding resistance R1, the quantities R1ia and R1ibare subtracted, or the subtraction of va-R1ia, vb-R1ib is performed. By a block 602', three-phase/ two-phase conversion is performed. By a block 603', the integrating circuit composition of a subtracting equation obtained from the block 601' is performed. From a block 650', the output of a primary magnetic-flux vector is generated.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention relates to a torque flux control method for an induction motor using a PWM (pulse width modulation) inverter, and in particular to an instantaneous control of the magnetic flux and torque of an induction motor. The present invention relates to a method for selecting a switching pattern that optimizes the switching pattern, and relates to a system that utilizes an instantaneous magnetic flux vector obtained by directly detecting and integrating the phase voltage of an induction motor.

[Conventional technology]

The basic operation of the torque flux control method for induction motors according to the present invention is described in the paper entitled "New high-speed torque control method for induction motors based on instantaneous slip frequency control" in Journal of the Institute of Electrical Engineers of Japan, Vol. 106, No. 1, Page 9. has been done.

In this paper, the motor magnetic flux is defined as the motor input voltage detected and integrated within the control circuit. The inverter output voltage is determined so that the length of this magnetic flux vector follows the given magnetic flux command and the tip thereof draws a circular locus. Further, the torque generated by the R3 motor is calculated as a vector product of the magnetic flux and the motor primary current, and an inverter output voltage whose magnitude follows the given torque command is determined. As a result, the instantaneous values of magnetic flux and torque are controlled to be maintained within a predetermined error range, the output voltage of the inverter is constantly updated at high speed, and instantaneous control of magnetic flux and torque is performed.

The present applicant is Patent Application No. 6119228 rPWM
We are currently proposing an inverter output voltage detection method. This will be explained briefly with reference to FIGS. 2 to 7.

FIGS. 2 to 7 are shown to explain an example in which the technology described in the above-mentioned paper employs the output voltage detection method previously proposed by the present applicant.

That is, in FIG. 2, from the DC voltage source 1 to the positive bus 1a
Power is supplied to the induction motor 5 via the PWM inverter 3 and the negative bus 1b. The control circuit 6 processes the command and detected current and voltage signals, and generates an energization signal for the switching elements of the PWM inverter 3.

The PWM inverter 3 is composed of six arms each consisting of transistors and diodes connected in anti-parallel, and as shown in the figure, there are three changeover switches Su, Sv,
It can be expressed as Sw. This PWM inverter 3
Power is supplied from each output terminal to the induction motor 5 via current detectors 4u and 4v14w, and a voltage detector 2 is connected between the positive and negative buses on the DC side, and each phase current and each Phase voltage can be detected.

Here, the primary terminal voltage and current of the motor are respectively v
When l j is 11 and the secondary current is 12, the voltage q2
It is a vector representation of the axis-transformed quantity, for example, vl is d
If the axis component is Vldl and the q-axis component is Vlq, then Vl
"Vld + j"Q ・・・・・・・・・
・・・・・・・・・・・・・・・・・・・・・It is indicated by ■,
If + Ill is similarly defined. Furthermore, the 0 on the left side of the equation represents the case where both d and q axis components are O, and in the case of a squirrel cage rotor, the secondary voltage is 0 as shown below. Secondary winding resistance Lll; Primary inductance R2; Secondary winding resistance L22; Secondary inductance N1; Mutual inductance am represents the rotational angular velocity, and p represents a differential operator.

On the other hand, as a definition of magnetic flux, one blown flux φ1 is φ, "' L
o i+ +M iz ・・・・・・・・・・・・
・・・・・・・・・・・・・・・■, and by expanding the first line of formula ■, V+ = (LL+ + pLB)! ++pM i
Substituting the z expression ■ and rearranging it, we get the following.

vl −R1t1= pφ1 ・・・・・・・・・
・・・・・・・・・・・・・・・・・・ ■Required from 1 part of the book.

Each changeover switch Su, Sv, Sw is connected to the positive bus line l.
There are cases where it falls to the a side and cases where it falls to the shell fsib side, and it never takes an intermediate position. Assuming that the former is state 1 and the latter is state O, the output states of the inverter can be expressed in the following switch state variable table.

Switch state variable table Here, k is a number indicating the changeover switch state, and this 8
There is only the street. Said, Vd, V, is d, q2
The actual d is a switch state variable expressed as a two-component.

The q-axis component vldlVlq can be expressed as P7 by multiplying it by the voltage ■ of the DC voltage source 1 and the root (2/3).

In addition, FIG. 3 shows the above switch state variable table, and the numbers in parentheses next to vl are the changeover switch 811°Sv,
It shows the state of Sw, and represents a voltage vector that steps clockwise by 60° as k increases. Note that k=0 and ni=7 are called zero vectors, which coincide with the origin in the figure.

1 (=Q and 7) are the induction motors. The line voltages of No. 5 are all 0, indicating a three-phase short circuit mode. Also, U.

The reference axes of the v and w phases are calculated as =
Corresponds to the directions 1, 3, and 5.

Next, the instantaneous torque T is determined by equation (2) as the vector product of the primary magnetic flux φ1 and the primary current 11 in equation (2).

T=+61Xi1=φtdX lid x 11q−φ
1qXftd...””■However, φ1d+φ1q and t, d+11q are the primary magnetic flux φl and 1, respectively.
These are the respective components when the next current i1 is decomposed into the d and q2 axes.

Furthermore, blocks 601 and 603b in the control circuit 6
is to calculate the primary terminal voltage Vl from the states of the changeover switches Su, Sv, Sw and the voltage V of the DC voltage source 1 detected by the voltage detector 2, and is calculated from the switch state variable table and the formula (■). Ru.

Block 602 is a block that converts the three-phase current '111'V l 'W detected by the current detector 411+4V+4W into d and q22 components using the following equation.

The primary terminal voltage V! is multiplied by the winding resistance R, in block 604. The product of primary winding resistance R1 and primary current 11 is subtracted from .

Block 605 is for performing an integral calculation of the magnetic flux according to the formula (■), and the d and q axis components of the primary magnetic flux φ1 φtd + φ1
q is required.

At block 610, as shown in the magnetic flux state diagram of FIG. 4, the primary magnetic flux φ! The rotation angle θ in the direction of the hour hand with respect to the d-axis of the vector is 30° and 90° as boundary lines.

150°, 210°, 270°, 33000) 60”
A control flag fθ is generated as follows depending on which region the object belongs to.

-300≦θ<30°: fθ=I 30°≦θ<90°; fθ=■ 90°≦θ<150°: fθ=■ 150°≦θ<210°; fθ=■ 210'≦e <270 ': fθ=V270°≦θ
<330°; fθ=■ Block 606 is for calculating the magnetic flux vector length φ1 using the following equation.

φx=W-7・・・・・・・・・・・・・・・・・・・・・・・・■Figure 5 is the state control diagram of the hysteresis comparator, where the magnetic flux vector length φ1 is the magnetic flux command value. For φl, using the error limit φ, * Δφ * φφ! 2<φ1
〈φ1 + 2... Group... 1 group... Eyes open...
- Generates a control flag fφ for controlling to [phase]. That is, when the magnetic flux vector length φ1 increases and reaches the upper limit of φ1 + 2, a control flag fφ=-1 that commands demagnetization is generated, and the magnetic flux vector length φ1 decreases after a little while and reaches the lower limit. When φ12 is reached, a flag fφ=1 is generated which commands magnetization. Thus, the magnetic flux vector length φ
1 is controlled to draw a limit cycle in the direction of the arrow shown in the figure, but in reality, block 60
The magnetic flux vector length φ calculated by formula ■ in 6! is subtracted from the magnetic flux command value *φ1 in block 608, and control flags fφ-1, -1 are generated in block 611 according to the state control diagram of FIG.

Regarding FIG. 4, the limit cycle of the magnetic flux shown in FIG. 5 corresponds to the fact that the vector tip of the primary magnetic flux φ1 is controlled so that it always exists in the illustrated annular portion. . Further, the control flag fφ and the control flag fθ described above are combined, for example, fφ=1, f
Assuming that the control flag for θ=engine is set, the area is 13
Since it means the magnetization mode at 0'≦θ<300,
The primary voltage Ml vector to be integrated into one outflow bundle φl vector has an outward component of a circle, and only one of FIGS. 3 and 4 or pH = 1.2.6 is selected. There is a possibility that

Block 607 calculates the instantaneous torque T by calculating the vector product of both outputs of blocks 602 and 605 using the formula (2). In block 6091, instantaneous torque T is subtracted from torque command T, A control flag fr is generated in block 612 so that the difference between the command *T and the instantaneous torque T determined by equation (2) is kept within a predetermined victory margin limit.

Fig. 6 is a state control diagram of the three-value hysteresis comparator, in which the torque deviation T*-T is the upper limit) Tl (
'Tl>0) When J is reached, an acceleration mode control flag ft=1 is generated. When the electric motor is accelerated and the torque deviation reaches the lower limit value -ΔT2 ('T2>O), the zero vector mode control flag fτ=O is generated, the torque gradually decreases, and the torque deviation increases again until it reaches the upper limit value. When Tl is reached, the mode shifts to the acceleration mode, and a limit cycle is drawn that goes around the hysteresis loop in the upper half of FIG. 6 in the direction of the arrow. Expressing this in the time domain, the instantaneous torque T fluctuates as shown in the torque waveform diagram of FIG. 7, and reciprocates within a band of upper and lower deviations ΔT1+ΔT2 with the torque command T in between. Next, when the electric motor is performing regenerative braking, a hysteresis loop is drawn in the lower half of Figure 6, and the torque deviation is the negative lower limit - Δ
When T1 (ΔTt>O) is reached, a deceleration mode control flag ft--1 is generated. Thereafter, the limit cycle in the direction of the arrow can be repeated in the same way as when moving. Thus, block 612 outputs a control flag ff=1, 0, -1.

Block 613 is blocks 610, 611, 612
This is a block that determines the most suitable inverter output voltage for each combination of the three control flags fθ, fφ, fv output from control flags fθ, f
φ, ft control.

For example, as described above, the control flag fφ=1. In the case of fθ=■, if the voltage vector is expressed as Vl(k) according to k in the switch state variable table, the potential to be selected as the voltage vector is either 1.2.6. However, at this time, if the control flag fτ=1, then =2, that is, the output voltage vector v4(2) is selected as a vector having a clockwise rotating component.

If fτ=-1, vl(6) is selected, and if fr±O, the zero vector is selected, and vl(0) or vt(7) is selected.

The switching table shown below shows the values of the output voltage vector numbers for all combinations of the three control flags fφ, fθ, and ft. By referring to , a switching signal is sent to the PWM inverter 3, and instantaneous control of magnetic flux and torque is performed.

Switching table Note that the inverter frequency can be considered as the rotational speed of one blown bundle φ1 vector in FIG. 4, but this is not given from the outside, but is generated as a result of integrating the voltage vectors according to equation (2).

[Problem that the invention seeks to solve]

As mentioned above, we have explained in detail the control method based on the space vector method, which is completely different from the conventional vector control method for rust conduction motors. Switching that calculates the instantaneous generated torque by calculating the vector product of the current 11 and the 1-blowout bundle φl, and optimizes the torque response of the inverter, which is tabulated in advance, according to the deviation between this and the torque command *T. It's about choosing a pattern.

However, as a drawback of such a control method, one blowout bundle φl is
Changeover switch 811 + Sy t s in 03b
The primary terminal voltage v is determined from the state of w, the voltage V of the DC voltage source 1 detected by the voltage detector 2, the switch state variable table, and the formula (■).
1, the c+ and q axis components φtd + φIq are obtained by integral calculation using a block 6051. Due to the unbalance of the inverter main circuit and the windings of the induction motor 5, the primary terminal voltage v1 is outputted as an ideal voltage without delay and free from these influences.

However, in the case of low speeds and light loads, the voltage waveform produced by a PWM inverter is not a simple square wave, but has many increments and its width is varied, so the effects of unbalance of transistor elements and dead time are particularly large. Therefore, the voltage actually applied to the induction motor has a narrow width corresponding to the dead time, and the effective voltage decreases. In other words, the effective voltages of each phase become unbalanced due to unbalance of each element. Therefore, the actual blown-out flux in the motor is smaller than the calculated 1-blown flux φ1 due to the put time, and is distorted due to imbalance, so that the magnetic flux does not draw a normal circular trajectory. As a result, the electric motor cannot rotate stably in a low speed range, causing rotational ripples, and the torque also includes ripples, making smooth control impossible.

In order to improve this, there are measures such as adjusting the dead time of each transistor in the PWM inverter and eliminating variations in the windings of the motor up to the low frequency range.

However, such improvement measures lacked versatility and could not serve as a solution.

[Means for solving problems]

In view of the above-mentioned points, the present invention is designed to directly take in the phase voltage of the induction motor in order to eliminate the distortion of the calculated magnetic flux surge and the problems of unstable rotation. That is, by directly detecting the two phase voltages va and vb, one blowout bundle φa,
This is based on the technical idea that φb is calculated from the following equation (■').

Therefore, the one-blowout bundle of each phase of the formula (2'1) trowel motor is obtained by the addition and integration circuit.

Furthermore, the d and q axes of one discharged bundle φl are decomposed into two axes:
The d-axis component φd and the q-axis component φq are obtained by the following equations.As shown in the equation 2', the d-axis component φd and the q-axis component φq are obtained by an adder circuit.

[For production]

By utilizing the technical means described above, problems related to the inverter output voltage obtained by multiplying the detected value of the inverter DC input voltage as described above and the voltage pulse width signal from the switch and switch state tables can be solved. In particular, the voltage/current detection means and magnetic flux integrating circuit of the induction motor will be described below.

That is, the three-phase phase voltages va and vb of the induction motor are detected as phase voltages divided by a resistive voltage divider circuit having a three-phase pseudo-neutral point, and the current is detected by the current detector 1a.

1b is detected.

Now, if we consider one phase of a three-phase induction motor, as in formula By adjusting the magnitude of the secondary resistance drop and the ratio with the phase voltage, it is possible to configure an integral circuit that calculates the amount of magnetic flux that evaluates one blown flux generated in an actual induction motor using the formula It will be easily obtained.

Here, φ3. The method to calculate φd and φq from φb is as follows:
From the 3PI3 phase-2 phase conversion formula and the condition (φ□+φb+φC=O), the formula ■'' holds true, so φd can be obtained by multiplying φ8 by an in constant.

Therefore, with these φd and φq, one blowout bundle vector is expressed as (
φ1=φd+jφq) and can be applied to magnetic flux and torque control.

In this way, when attempting to detect the rising and falling timing of the inverter output voltage waveform directly from the voltage detector 2, a general voltage detector has a large delay and cannot correctly detect the steep inverter output voltage. However, according to this method, the rise of the inverter output voltage waveform.

Unlike detecting the fall timing, we use a system that handles the integral amount of the voltage detection amount, so the offset of the control deviation is removed and the integrator output l is fed back through an appropriate gain, so the Compared to the method described in the prior art section, much finer control can be achieved.

Further, the present invention will be explained in detail with reference to embodiment drawings.

〔Example〕

FIG. 1 shows the main configuration of an embodiment to which the present invention is applied, in which 6' and 6'' are control circuits, and 7u, 7v, and 7w are voltage detectors. The same symbols indicate parts having the same function.

That is, in the system shown in FIG. 1, a voltage detector 70t 7V * 7W is added to the system shown in FIG.
, 6// are provided. Here, the control circuit 6'' includes blocks 601 and 602 of the control circuit 6.
.

603a, 603b, 604, and 605 have been removed.

In the control circuit 6', a block 601' uses voltage detectors 7u, 7v, and 7w to detect phase voltages VB, Vl) from a three-phase voltage divider circuit with a pseudo-neutral point, and a current detector 4u*4v. The resistance drop due to the primary winding resistance R was considered in the phase currents ia and ib detected at +4W (Rtia).

From the detected amount (R1[b), (vB R, 1tB
) + (vb-Rlib). As the (R, I) compensation, the magnitude of R1 is the amount of voltage drop here, so if the value of R1 is set as voltage gain adjustment, it is possible to adjust the manipulated variable of resistance drop with respect to voltage and current. be.

Block 602' is for performing three-phase to two-phase conversion, and block 603' is for constructing the subtractive type integration circuit obtained in block 601'.

Blocks 604a' and 604b' act as isolators to isolate the electrical potential of the motor charged up to blocks 603' and 602' from the output signal to the next stage. It goes without saying that the conversion function of block 602' may be performed after block 604b'.

As mentioned earlier in the section of action, Pro 605' is
This is an addition and gain constant adjustment section that converts φa, φb into φd, φq, and φd, φq decomposed into d-q axes.
Since this is obtained, one outflow bundle vector is (φ1=φd+jφq).

The control circuit 6'' is composed of the above-mentioned components, which have been explained in connection with the control circuit 6 shown in FIG. 2, so detailed explanation thereof will be omitted here.

〔Effect of the invention〕

In a torque flux control method for an induction motor in which torque can be calculated by the vector product of the primary magnetic flux expressed as an instantaneous space vector and the primary current, the primary magnetic flux is obtained by integrating the motor terminal voltage.

Therefore, by detecting the output voltage and current of the inverter, the magnetic flux and torque can be controlled simultaneously and independently, without using constants or variables on the secondary side such as vector control as mentioned above. However, among the voltage detection means for obtaining the primary magnetic flux, which is the main control variable in this torque control system, there is a delay in the output voltage due to the dead time of the switching transistor, and the main circuit and motor windings. When the influence of the three-phase voltage unbalance associated with the impedance unbalance becomes large, the primary magnetic flux that integrates the voltage will inevitably become unbalanced, so the rotational angular velocity of the Lissajous magnetic flux will be constant. However, such problems can be improved, especially in low speed ranges with a speed ratio of (100:1) or less or under light loads.

As explained above, according to the present invention, it is possible to control the voltage distortion due to the influence of de-time at low speeds and light loads, and the unbalance of the main circuit mentioned above, so that the normal rotation of one blown bundle and perfect roundness can be achieved. It is possible to provide an exceptional system that has remarkable practical value, which provides a high adhesive surge, a low-speed stable circuit, and a reduced torque ripple.

[Brief explanation of drawings]

Fig. 1 is a schematic diagram showing the main part configuration of an embodiment to which the present invention is applied, and Figs. Figure 2 is a block diagram showing the control system, Figure 3 is a vector diagram of the inverter output voltage depending on switch state variables, Figure 4 is a magnetic flux state diagram, and Figure 5 is a diagram of the magnetic flux state. FIG. 6 is a state control diagram of a binary hysteresis comparator for magnetic flux, FIG. 6 is a state control diagram of a three-value hysteresis comparator for torque, and FIG. 7 is a torque waveform diagram of an electric motor. l...DC voltage source, 2,7ur7v*7w...
...Voltage detector, 3...PWM inverter, 411 r 4v * 4w... Current detector, 5... Induction motor, 6, 6', 6''・
...Control circuit.

Claims (1)

[Claims] The instantaneous values of magnetic flux and torque are calculated from the voltage and current signals of the induction motor, and the deviations between the calculated magnetic flux and torque and the respective command values are calculated within a predetermined tolerance to the given magnetic flux and torque commands. Similarly, in a torque flux control method for an induction motor using a PWM inverter that selects a switching pattern that optimizes the torque response of an inverter that is tabulated in advance, the voltage signal is directly detected by the phase voltage of the induction motor. In addition, it also detects the primary resistance voltage drop for one phase, adds and integrates the detected value, and obtains the primary magnetic flux amount as an integral amount, thereby generating a primary magnetic flux vector and performing instantaneous torque control. A torque flux control method for an induction motor, which is characterized by: